These designs take the en-network detection approach: misbehaved nodes are detected by their neighboring watchdog nodes. The final decision of whether or not a suspicious node is compromised is determined by a group voting procedure. Li, Song and Alam have defined a data transmission quality function which keeps close to constant or change smoothly for legitimate nodes and decreases for suspicious nodes. Hinds has used Weighted Majority voting algorithm to create a concept of a node which could not be compromised, and to develop detection algorithms which relied on the trustworthiness of these nodes. Wang & Bagrodia have designed an intrusion detection system for identifying compromised nodes in wireless sensor networks using common application features (sensor readings, receive power, send rate, and receive rate). Hence we provide the literature review in two parts, first, how compromised nodes are currently filtered and second, on the use of AR-HMM for solving diverse problems of identification, filtering, and prediction. However detecting compromised nodes using AR-HMM is a new area of investigation and we were unable to find any references that researched the same. Baum and his peers at the Institute for Defense Analysis in the late 1960s.
#Hidden markov model matlab forecasting series
Hidden Markov models and the Baum-Welch algorithm were first described in a series of articles by Leonard E. In Section IV, we describe the proposed algorithm, and in Section V we give numerical results. In Section III, we give a brief overview of AR-HMMs. In the next Section II we cover the literature review for this work. In this paper, we propose a method to filter the compromised nodes, be it self-healing or corrupted, using an autoregressive hidden Markov model (AR-HMM). Anomaly detection is a key challenge in ensuring both the security and usefulness of the collected data. However, sensor networks are vulnerable to adversaries as they are frequently deployed in open and unattended environments. Sensor network deployment is becoming more commonplace in environmental, business and military applications. For example, the sensor nodes in a wireless sensor network can be used collaboratively to collect data for the purpose of observing, detecting and tracking scientific phenomena. Sensor systems have significant potential for aiding scientific discoveries by instrumenting the real world.
Keywords: Autoregressive hidden Markov models environment sensing filtering corrupted nodes sensor network clustering anomaly detection Simulations using both synthetic and real datasets show greater than 90% accuracy in identifying healthy nodes with ten nodes datasets and as high as 97% accuracy with 500 or more nodes datasets. Our approach is a simple, decentralized model to identify compromised nodes at a low computational cost. The existing algorithms are centralized and computation intensive. For each node, we train an AR-HMM based on the sensor's readings, and subsequently the B matrices of the trained AR-HMMs are clustered together into two groups: healthy and compromised (both self-healing and corrupted), which permits us to identify the group of healthy sensors. A different AR-HMM ( A, B, π) is used to describe each of the three types of nodes. We assume that sensors are healthy, self-healing and corrupted whereas each node submits a number of readings. We propose a method based on autoregressive hidden Markov models (AR-HMM) for filtering out compromised nodes from a sensor network.