Mobile Edge Computing (MEC) has been gaining a major interest for use in Cyber-Physical Systems (CPS) for Disaster Response and Tactical applications. These CPS generate a very large amount of mission-critical and personal data that require resilient and secure storage and sharing. In this article, we present the design, implementation, and evaluation of a framework for resilient data storage and sharing for MEC in CPS targeting the aforementioned applications. Our framework is built on the resiliency of three main components: EdgeKeeper, which ensures resilient coordination of the framework’s components; RSock, which provides resilient communication among CPS’s nodes; and R-Drive/R-Share which, leveraging EdgeKeeper and RSock, provides resilient data storage and sharing. EdgeKeeper employs a set of replicas and a consensus protocol for storing critical meta-data and ensuring fast reorganization of the CPS; RSock decides an optimal degree for replicating data that is communicated over lossy links. R-Drive employs an adaptive erasure-coded and encrypted resilient data storage; R-Share, leveraging RSock provides resilient peer-to-peer data sharing. We implemented our proposed framework on rapidly deployable systems (e.g., manpacks, testMobile Edge Clouds) and on Android devices, and integrated it with existing MEC applications. Performance evaluation results from three real-world deployments show that our framework provides resilient data storage and sharing in MEC for CPS.

In this paper, we show that Hybrid Routing and Prophet protocols in Opportunistic Mobile Networks (OMNs) are vulnerable to theCollusiveHijackattack, in which a malicious attacker, Eve, compromises a set of nodes and lies about their Inter-Contact-Times (ICTs). Eve claims that her nodes meet more frequently than in reality to hijack the routes of legitimate nodes in OMNs. The CollusiveHijack attack enables Eve to launch more severe attacks like packet modification, traffic analysis, and incentive seeking attacks. To identify the CollusiveHijack attack, we propose the Kolmogorov-Smirnov two-sample test to determine whether the statistical distribution of the packets’ delays follows the derived distribution from the ICTs among the nodes. We propose three techniques to detect the CollusiveHijack attack, the Path Detection Technique (PDT), the Hop Detection Technique (HDT), and the Early Hop Detection Technique (EHDT), which trade off compatibility with the Bundle Security Protocol, the detection rate, and the detection latency. We evaluated our techniques through extensive trace-driven simulations and a proof-of-concept system implementation and show that they can detect CollusiveHijack attacks with 80.0% to 99.4% detection rates (when Eve hijacks more than 60 packets) while maintaining a low false positive rate ($\sim$3.6%) and a short detection latency (7-14 hours) for EHDT (75%-85% enhancement compared to PDT and HDT).

Mobile Edge Computing (MEC) is gaining significant interest from first responders and tactical teams. Typical cloud-based coordination (e.g., service discovery and coordination, device naming, authentication) does not work in MEC due to high user mobility. We design and implement EdgeKeeper to provide cloud-like service coordination to distributed edge computing applications for MEC systems. It maintains an edge cluster among devices and intelligently stores data on a group of replicas to guard against node failures/disconnections. We provide a full-system implementation of Edge-Keeper for Android and Linux platforms and evaluate it with MEC applications in a real-world wide-area search and rescue operation conducted by first responders.

In this paper, we first show that Group Owner (GO) devices in Wi-Fi Direct are vulnerable to the EvilDirect attack. In the EvilDirect attack, a rogue GO is set up by an adversary to look like the legitimate GO (with the same MAC address, SSID, and operating channel). The adversary intercepts the clients' invitation requests and accepts them before the legitimate GO. Accordingly, the adversary hijacks the wireless communications between the clients and the legitimate GO. To defend against the EvilDirect attack, we propose the idea of exploiting the received signal strength (RSS) variations on the wireless channel between each client and the legitimate GO. Our solution, EvilDirectHunter checks whether the RSS profiles of both the client and the potential GO devices are similar with each other. Both devices incrementally prove this similarity by exchanging challenge and response packets. EvilDirectHunter is evaluated by implementing it as an Android App, and by modifying the Android kernel code responsible for Wi- Fi Direct in Google Nexus 5 and Samsung Galaxy S2 smartphones. The results show that EvilDirectHunter is able, within seconds, to identify EvilDirect attacks with a high detection rate (100%) while maintaining a low false positive rate (4.5%).

In this paper we first show that the Wi-Fi Protected Setup (WPS) protocol (used by Wi-Fi Direct, the de facto adhoc communication mechanism for smartphones and mobile devices) is vulnerable to a brute-force or dictionary attack. To defend against these attacks, we propose the idea of using contextual information (i.e., data obtained from mobile device's sensors) to establish a long (128 bits) secure session key between two Wi-Fi Direct enabled devices, instead of using the keypad. Our solution, Session Key Generated from Sensors (SekGens) employs three phases. In the Quantization Phase, the key is iteratively generated based on different sensors' data. In the Reconciliation Phase, the two devices eliminate minor differences in the bits of their keys by using the Cascade reconciliation mechanism. In the Privacy-Amplification-and-Hashing Phase, the two devices omit all bits exposed during the reconciliation phase and apply hashing to the remaining secret bits. SekGens is implemented and evaluated by modifying the Android kernel code responsible for WPS in Google Nexus 5 and Samsung Galaxy S2 smartphones. The results show that SekGens generates keys with low mismatch ratio (less than 3%), at a fast rate (~20 bits/sec), and with high entropy (~92%).

An Ad-hoc Mobile Cloud (AMC) is a new computing model that allows sharing computing power of multiple mobile devices. For a diverse group of individuals that employ such computing model, in an ad-hoc manner, secure peer-to-peer communication becomes very important. Using private or pairwise keys to secure such communication is preferable to public-keys because of computation and energy requirements [1]. With the advent of sensor enabled mobile devices, a protocol (SekGens) that uses sensor data to generate pairwise keys on demand has been proposed [2]. To work successfully SekGens requires devices to be closely located and becomes infeasible for devices situated multiple hops away. SekGens is also expensive in computation and slow in key generation. In this paper, we investigate how to enable devices in an AMC to establish pairwise keys. We propose an efficient solution which tries to reduce the number of executions of SekGens in the AMC, and establishes pairwise keys between nodes multiple hops away by distributing parts of the key on multiple routing paths. Our results show a reduction of up to 75% in the number of SekGens required to establish keys in an AMC, when compared to a naive approach.