Content Dissemination Over VANET: Boosting Utility Based Heuristics Using Interests

Introduction:

This paper focuses on dissemination of decentralized opinion polls over VANETs (Vehicular Ad-hoc Networks) of self-interested peers. For comparing heuristics, we compute utility of achieved dissemination from sender’s perspective. The main goal is to calculate behaviour of self-interested senders. Utility model is also generated where the highest is given to the items generated by sender followed by similar opinion and least is given to opposite opinion. Based on VANET experiments we determine parameters for the model.

Theoretical and Empirical issues:

In this paper, there are a few logical inconsistencies. Like on page 1, he failed to give more insight about the “DV-CAST” broadcasting protocol and the way he justified it by giving a reason due to increase in performance is disappointing. Some of the measures used in this study were validated by past studies but when placed together there are overlaps among the measures. The author also did a logical assumption that in uninformed heuristics all peers are interested in all the items current peer has, he did not give enough justification to prove his theory. If he did so, then it would give a good understanding to the reader. He also mentioned that the communication model presumes that each device broadcast data which it wants to disseminate at the same time it processes and listens to data that is broadcasted by nearby devices. But he could have also given more explanation for this, I suspect that even other readers would expect this in the paper.

Assumptions:

In this paper in some instances the author has given unexplained assumptions that need more clarification. I stated some of them below.

1. In page 1, each device is owned by user and the system is assumed to be open which means that user has complete control over the device.

1. In page 2, Software agents on wireless ad-hoc devices are assumed to broadcast messages consistently.

1. In page 4, queues that are used for sending data are long enough to supply data for the whole time and

1. In page 5, the author assumed that Np<<No but he did not give any evidence to prove his point.

Positives:

I appreciate the great work done by the author and I mentioned some of them below.

1. In page 3, simpler algorithms are used to handle received vote and witness messages it also carries other items.

1. In page 6, in the three scenarios the number of messages is successfully transmitted, and they are also shown in a proper tabular format.

Negatives:

I mentioned some of the drawbacks of the paper below on which the author should work on.

1. In page 2, In DirectDemocracyP2P, peer agents only store last signature given by each constituent for each motion.

1. In page 4, the author included some mathematical formulas, but he did not explain them clearly.

1. In page 6, the author mentioned that configuration of the network is automated on Windows and Linux but while on MacOS it must be done manually.

Overall, this the selected topic is interesting and despite its flaws it offers informative data to the readers.

pur-new-sol

Related Questions