Data aggregation has been one of the most widely researched topics for Wireless Sensor Network (WSN) as it helps in increasing the effective life time of the network and reduces the packet transmission overhead. Besides other challenges, security is one of the serious challenges during data aggregation. Numerous solutions on security for data aggregation for WSN are based on homomorphic encryption technique or Elliptic Curve Cryptography (ECC) based solutions and thus are vulnerable to attacks by quantum machines. This paper presents a post-quantum security solution using lattice cryptography. The proposed solution is applying Learning with Errors over Rings (R-LWE) for encryption of data in the data aggregation process at the gateway module of WSN. Security analysis shows that the proposed scheme provides confidentiality, integrity and authenticity during communication. Performance analysis shows that the proposed scheme is lightweight and shows better performance compared to Elliptic Curve Elgamal (ECEG) cryptography based scheme and symmetric homomorphic based scheme.

Privacy preservation is one of the important research challenges in IoT applications. In one such IoT application; smart grid network, billing information and energy profiling information of the customer may be collected, aggregated, and forwarded to control centre for further analytics. Based on the research findings, traditional public key cryptography is not secured against quantum attacks. Our study is motivated by the recent developments in the lattice-cryptography schemes. This paper presents a lightweight R-LWE lattice-cryptography-based scheme to sign and encrypt message traffic in smart grid. Security analysis suggests that proposed scheme preserves the privacy of customer. Performance analysis shows that proposed scheme cause less communication overhead as compared to traditional public key cryptography yet maintain parallel with NTRU-based scheme and outperforms both formats of public key cryptography in regards to computation overhead.

Data migration has become a priority in many industries, spawned by a variety of business needs. Most of the existing tools for Extract, Transform and Load (ETL) process of data migration are piece-meal and do not present a complete solution. Moreover, while research has focused on the problem of Schema Mapping, a key step in the ETL process, most of the current algorithms do not perform well on real-world data. Researchers have suggested the use of Domain Knowledge to enhance schema mapping. In this paper, we use domain knowledge in an innovative manner to improve schema mapping in an 'actual' industrial setting. Further, we take a comprehensive view of the data migration problem and present an end-to-end system for the ETL process, utilizing existing tools for each step and building connectors, wherever required. We focus on Data Migration for Telecom Billing and utilize domain knowledge captured in an ontology, a thesaurus and a set of rules to improve schema mapping. Experiments conducted on a real-life data demonstrate the effectiveness of our system and validate the utility of domain knowledge in data migration projects.