Quality of service (QoS) is the ability of a set of networked resources--such as network elements, infrastructure devices, host computers, or a myriad of applications--to deliver traffic with minimum delay and maximum availability. Recently, QoS included two elements--secure QoS (or QoS that is secure) and security service QoS--that are services offered against service-level agreements (SLAs). This paper attempts to put these two elements together for pervasive systems and networking, but fails to do so.

The main confusion in the paper arises with the repetition of buzzwords and terms, and the lack of accuracy and clarity across several subject domains, such as software engineering, security, autonomic computing, and algorithm analysis. In addition, the keywords are inconsistently used. Difficulty in understanding arises again when one considers other parts of the proposed model, ranging from subjective to primitive objective parameters in the context of multiple constraints, utility-based use of computing resources, perceived risks, and user preferences.

This model--an extension of previous work by one of the authors--includes “perceived risk level and limitations in computing resources.” The authors offer minimal tangible output or evidence that their system can actually work in a real environment. I would not recommend this paper for understanding QoS.