Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDistPrivacy: Privacy-Aware Distributed Deep Neural Networks in IoT surveillance systems
246
0
0.0
(
0
)
Added by
Emna Baccour
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
With the emergence of smart cities, Internet of Things (IoT) devices as well as deep learning technologies have witnessed an increasing adoption. To support the requirements of such paradigm in terms of memory and computation, joint and real-time deep co-inference framework with IoT synergy was introduced. However, the distribution of Deep Neural Networks (DNN) has drawn attention to the privacy protection of sensitive data. In this context, various threats have been presented, including black-box attacks, where a malicious participant can accurately recover an arbitrary input fed into his device. In this paper, we introduce a methodology aiming to secure the sensitive data through re-thinking the distribution strategy, without adding any computation overhead. First, we examine the characteristics of the model structure that make it susceptible to privacy threats. We found that the more we divide the model feature maps into a high number of devices, the better we hide proprieties of the original image. We formulate such a methodology, namely DistPrivacy, as an optimization problem, where we establish a trade-off between the latency of co-inference, the privacy level of the data, and the limited-resources of IoT participants. Due to the NP-hardness of the problem, we introduce an online heuristic that supports heterogeneous IoT devices as well as multiple DNNs and datasets, making the pervasive system a general-purpose platform for privacy-aware and low decision-latency applications.
rate research
Read More
Wireless sensor networks (WSN) are fundamental to the Internet of Things (IoT) by bridging the gap between the physical and the cyber worlds. Anomaly detection is a critical task in this context as it is responsible for identifying various events of interests such as equipment faults and undiscovered phenomena. However, this task is challenging because of the elusive nature of anomalies and the volatility of the ambient environments. In a resource-scarce setting like WSN, this challenge is further elevated and weakens the suitability of many existing solutions. In this paper, for the first time, we introduce autoencoder neural networks into WSN to solve the anomaly detection problem. We design a two-part algorithm that resides on sensors and the IoT cloud respectively, such that (i) anomalies can be detected at sensors in a fully distributed manner without the need for communicating with any other sensors or the cloud, and (ii) the relatively more computation-intensive learning task can be handled by the cloud with a much lower (and configurable) frequency. In addition to the minimal communication overhead, the computational load on sensors is also very low (of polynomial complexity) and readily affordable by most COTS sensors. Using a real WSN indoor testbed and sensor data collected over 4 consecutive months, we demonstrate via experiments that our proposed autoencoder-based anomaly detection mechanism achieves high detection accuracy and low false alarm rate. It is also able to adapt to unforeseeable and new changes in a non-stationary environment, thanks to the unsupervised learning feature of our chosen autoencoder neural networks.
Age of Information (AoI) has gained importance as a Key Performance Indicator (KPI) for characterizing the freshness of information in information-update systems and time-critical applications. Recent theoretical research on the topic has generated significant understanding of how various algorithms perform in terms of this metric on various system models and networking scenarios. In this paper, by the help of the theoretical results, we analyzed the AoI behavior on real-life networks, using our two test-beds, addressing IoT networks and regular computers. Excessive number of AoI measurements are provided for variations of transport protocols such as TCP, UDP and web-socket, on wired and wireless links. Practical issues such as synchronization and selection of hardware along with transport protocol, and their effects on AoI are discussed. The results provide insight toward application and transport layer mechanisms for optimizing AoI in real-life networks.
In the last decade, many semantic-based routing protocols had been designed for peer-to-peer systems. However, they are not suitable for IoT systems, mainly due to their high demands in memory and computing power which are not available in many IoT devices. In this paper, we develop a semantic-based routing protocol for dynamic IoT systems to facilitate dynamic IoT capability discovery and composition. Our protocol is a fully decentralized routing protocol. To reduce the space requirement for routing, each node maintains a summarized routing table. We design an ontology-based summarization algorithm to smartly group similar capabilities in the routing tables and support adaptive routing table compression. We also design an ontology coding scheme to code keywords used in the routing tables and query messages. To complete the summarization scheme, we consider the metrics for choosing the summarization candidates in an overflowing routing table. Some of these metrics are novel and are difficult to measure, such as coverage and stability. Our solutions significantly reduce the routing table size, ensuring that the routing table size can be bounded by the available memory of the IoT devices, while supporting efficient IoT capability lookup. Experimental results show that our approach can yield significantly lower network traffic and memory requirement for IoT capability lookup when compared with existing semantic-based routing algorithms including a centralized solution, a DHT-based approach, a controlled flooding scheme, and a cache-based solution.
Geo-distributed private chain and database have created higher performance requirements for consistency models. However, with millisecond network latency between nodes, the widely used leader-based SMR models cause frequent retransmission of logs since they cannot know the logs replication status in time, which resulting in the leader costing high network and computing resource. To address the problem, we proposed a Leader Confirmation based Replication (LCR) model. First, we demonstrate the efficacy of the approach by designing the Future-Log Replication model, which the followers are responsible for non-transactional log replication. It reduces the leaders network load using the signal log. Secondly, we designed a Generation Re-replication strategy, which can ensure the security and consistency of future-logs when the number of nodes changes. Finally, we implemented LCR-Raft and designed experiments. The results show that in the single-ms network latency environments, LCR-Raft can provide 1.5X higher TPS, reduces transactional data response time 40%-60%, and network traffic by 20%-30% with acceptable network traffic and CPU cost on followers. Besides, LCR can provide high portability since it does not change the number of leader and election process.
In this demonstration, we present a privacy-preserving epidemic surveillance system. Recently, many countries that suffer from coronavirus crises attempt to access citizens location data to eliminate the outbreak. However, it raises privacy concerns and may open the doors to more invasive forms of surveillance in the name of public health. It also brings a challenge for privacy protection techniques: how can we leverage peoples mobile data to help combat the pandemic without scarifying our location privacy. We demonstrate that we can have the best of the two worlds by implementing policy-based location privacy for epidemic surveillance. Specifically, we formalize the privacy policy using graphs in light of differential privacy, called policy graph. Our system has three primary functions for epidemic surveillance: location monitoring, epidemic analysis, and contact tracing. We provide an interactive tool allowing the attendees to explore and examine the usability of our system: (1) the utility of location monitor and disease transmission model estimation, (2) the procedure of contact tracing in our systems, and (3) the privacy-utility trade-offs w.r.t. different policy graphs. The attendees can find that it is possible to have the full functionality of epidemic surveillance while preserving location privacy.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
