Scalable, Secure and Resilient Cloud Platform for Centralised Control, Management and Monitoring
Cloud scalability, elasticity and security are critical features for adoption of cloud. Development of a user-centric cloud that provides more control and visibility for continuous monitoring to the cloud user is the objective of this project. Some of the key challenges researched include: stability, reliability and encapsulation of typical complexities associated with hypervisors of this Amrita Cloud platform.
Formal Models for Trust in Cloud Computing
In the cloud computing environment trust model helps to find more trustworthy providers for cloud hosting, which indirectly helps a cloud consumer to compute the degree of trust one can instill on a cloud provider. Addressing security and privacy issues in cloud computing alleviates the concerns of tenants. Such a model also adds value to the technological advancements in computing.
The two types of approach being researched are: the logical approach and quantitative approach. Logical approach is based on trust's semantic structure and its logical condition, whilst quantitative approach is based on uncertainty of trust, trust qualification, trust dynamics, and trust computations models and algorithms.
Internet of Things and Security
The Internet has evolved over the years from a network of computer clusters to one that also encompasses heterogeneous devices capable of generating and consuming data. This expanded network is referred to as an 'Internet of Things' (IoT). There is no clear definition that would mark the boundaries of its operational spectrum.
From its initial concept supporting RFID devices, IoT has broadened its scope to include road traffic monitoring devices, building surveillance cameras, home utility-metering devices to personal medical devices. In short, the IoT spans nearly every walk of life and is clearly the way of the future Internet.
This research addresses the unique naming of devices for identification, secure device registration and recognition. It also involves designing a new programming language to handle concurrent processing of events generated by participating devices.
Boolean functions play a crucial role in numerous cryptographic algorithms (for example in stream ciphers), and the study of their properties is therefore of great importance. Our research focuses on designing algorithms for the classification of Boolean functions. We are particularly interested in algorithms computing the higher orders non-linearities of Boolean functions. Investigations of list decoding algorithms for Reed-Muller codes and their applications to cryptanalysis is being pursued.
Health care security has evolved to become one of the most prominent areas of Cyber Security. With the increasing concerns about the recent exploitation of the security vulnerabilities of the medical devices, which potentially risks human lives, the need to invent secure, smart and low cost medical devices and eco-system has increased. This fact is exacerbated due to the rising number of old age population with chronic illness, and the dramatically rising cost of healthcare and its mandated compliance. This has inspired Amrita Cyber Security center to undertake research in the area of health care security with a focus on the following health care security research areas.
- Secure remote health care using Mobile communications and Cloud technologies.
- Security of the connected medical devices and systems.
- Wearable and Implantable medical devices (IMD) security.
- Privacy and security of the health care data
BigData Analytics Platform
IoT systems generate large amount of data and it demands a big data platform to effectively store and methodically analyse the streaming data generated by participating devices for real-time analysis and reporting. In most cases these decision supportive intelligence should be instantaneous.
Security challenges in this open platform are greater because the environment provides a shared file system among all platform users. This is unlike the conventional big data infrastructure offered by traditional cloud service providers where dedicated servers and processing file systems are offered even though it is being hosted in a cloud environment.
Key areas of research focus are;
- Big Data platform with distributed parallel data bases which is auto-scalable and highly available.
- Predictive data analysis and auto-alert management and monitoring.
- Multi source data adaptors which can work with different IoT input devices or systems.
- Built-in Data mining engines which performs incremental and rule based data mining to bring out intelligence from the data streams.
- Customisable mining rule maps that will grow as per the intelligence it gathers using machine learning techniques or algorithms.
- Parallelising several of the well known algorithms to run in Hadoop and Spark Clusters
The objective of this research is to build a new Automotive Infotainment Platform that will facilitate and secure reliable inter-vehicular data exchange. This Platform will also have the capability of segregating the internal and external networks using a built-in security layer. The above system will also assist in reducing traffic congestion in Indian Roads by sensing in advance possible traffic build ups and detecting traffic congestions ahead thereby advising drivers to take alternative routes. Research is underway to build other supporting low cost road side traffic monitoring systems to assist the task.