The Wireless Sensor Networks (WSNs) protocols commonly use in-network processing to optimize the communication costs. In-network processing involves processing of the sensed data on-the-fly during the course of the communication to the base station. However, due to the fusion of data items sourced at different nodes into a single one, the security of the aggregated data as well as that of the aggregating node, demands critical investigation. There has been numerous proof-of-concept attempts published in the literature, that devise secure data aggregation protocols offering either end-to-end or hop-by-hop secure aggregation. However, as per our modest observations, an integrated framework that offers all the necessary security attributes viz. confidentiality, privacy and integrity for secure data aggregation in WSNs is required. In this paper, we propose a framework for privacy preserving secure data aggregation in WSNs that provides support for confidentiality, privacy and integrity collectively. To support the confidentiality and privacy attributes in the framework, we evaluate the classical homomorphic encryption algorithms by comparative evaluation. Subsequently, based on this elaborate evaluation, we integrate the optimal algorithm amongst these, into tree topology in WSNs. To support the data integrity attribute, we analyse Elliptic Curve Digital Signature Algorithm (EC-DSA) and all of its published variants. However, in order to enhance the security strength of the basic EC-DSA, we propose our own variant of EC-DSA. In our humble observations, ours is a unique attempt that integrates the support for confidentiality, privacy and integrity in defined tree architecture for secure data aggregation protocol. We substantiate the proposal with elaborate experimental evaluation, too.