Secured Anomalous Detection for Personal Healthcare System

Author Name(s): Dr. Pradeep. B. Mane, Divya Katariya
Author Email: pbmane6829@rediffmail.com

Abstract

Advances and developments in medical systems have prompted a heightening in Implantable and Wearable Medical Devices(IWMD’s).IWMD’s normally incorporate remote correspondence interface through which they can be associated with Body Area Network(BAN). However, the programmability and remote availability of medicinal gadgets open up open doors for malicious attackers otherwise called anomalies. Therefore, there is a need to detect anomaly which is a vital issue that has been explored inside diverse research zones and applications. In this paper, a non-invasive method or secured framework is used for detection of anomaly that fortify against Radio Frequency attacks that are wirelessly caused on the PHS’s(Personal Healthcare System). The proposal is based on a medical security monitor that investigates all the radio-frequency wireless communications to/from medical devices. It identifies malicious transactions using 3 techniques viz, RSSI (Received Signal Strength Indicator), Password and Time Anomaly. Upon detection of an anomalous transaction, it takes suitable response actions such as jamming the commands so that they do not harm the therapeutic gadget. The performance of the system is determined by developing a prototype implementation for monitoring Lung Capacity using Spirometer along with Temperature monitoring using two Radio Peripheral boards i.e. Zigbee. Moving towards the 3rd layer of protection, proves that the system is able to obstruct all kind of attacks resulting in 100 % accuracy showing that it is an effective method to increase the security of BAN.

Introduction

Medicinal advances and in addition developments in ultra-low-control registering, systems administration, and detecting innovations have led to an increase in wireless accessible implantable and wearable medical devices(IWMDs). IWMD’s can communicate directly with the server and other devices in the network through remote correspondence interfaces. These remote correspondences include sensors, actuators that are connected to the BAN. A PHS(Personal Healthcare System) can be defined as a network of communicating devices such as sensor or actuators that can provide health information to the user of providing remote health services to the user. The functions performed by the wireless accessible medical devices are sometimes found to be life-critical. So, there is a need to check for any malfunction in their operation. Hence, there is a need for development of system that will allow for the detection of such malfunction/abnormalities or otherwise called Anomalies. Anomaly Detection is defined as detection of abnormalities in the data that do not match with the normal data. Anomaly Detection finds wide range of applications such as detection of fault for cyber security, intrusion detection for health care, fraud detection for credit cards, etc. In this paper, overviews of the attacks are described. To protect the medical devices against any such kind of wireless attacks, three layered Anomaly Detection Technique is described in this paper. The attacker can eavesdrop on the data packets wirelessly send by the remote control. The attacker can try to get the device PIN from the captured or transmitted packets. By changing the settings of the remote control, the attacker can harm the patient by disabling or changing the intended therapy. The false data sent by the attacker can reach the monitor or the controller so that the system provides wrong or unintended therapy to the patient. It is hard to distinguish the assailant’s transmissions from the genuine ones. Secured Anomalous Detection for Personal Healthcare System detects wireless attacks using three layered security. It is based on the observation of the physical characteristics of the signal, password/PIN that is embedded in the packet and the time during which the signals are sent. The attacker’s password/PIN may match or conform to the communication protocol, they may vary in the physical signal characteristics from the legitimate transmission. The framework utilizes 3 layered peculiarity identification technique to track every single malignant exchange. Upon detection of an anomaly or malicious activity, the monitor jams the data sent by the anomaly.

Conclusion

In this paper, a non-invasive method or secured framework is used for detection of anomaly that fortify against Radio Frequency attacks that are wirelessly caused on the PHSs. This secured framework solution aims at increasing the reliability of the system and thereby increasing the patient’s integrity and data confidentiality. Using 3 layered anomaly detection technique; it detects all the wireless communication caused to the medical devices/from the medical devices identifying potentially malicious transactions. This system is able to detect 66% of attacks using1st layer of security i.e RSSI. The 1st layer is not 100% reliable and hence there is a need for 2nd and the 3rd layer of security. It is possible that a skeptical attacker may evade the 1st layer of security by knowing the physical characteristics of the signals, the other types of attacks can be caught by switching to Password/PIN anomaly. The 2nd layer of anomaly is 98% reliable stating that it is however better than the 1st layer of security. Moving towards the 3rd layer of security i.e Time Anomaly, the system is able to detect all attacks resulting in 100 % accuracy.

689 total views, no views today

Download PDF File

About the author: admin