Zhu et al. propose “a novel, secure architecture to enable secure communications in large-scale, multi-domain wireless mesh networks.” According to the authors, the “mesh clients can conveniently gain services securely, even when roaming.” Overall, the basic idea seems good.

This paper covers basic Internet-based computing (IBC) and the authors’ contribution toward enabling secure communications. It reads like a conference paper in that it lacks the depth and details typically present in a journal paper. It adopts an approach that is mostly theoretical.

A significant weakness is that the paper lacks experimental results from simulation, actual experiments, or mathematical analysis. For example, in Section 3.2, the authors state:

One serious concern one may have in using ID-based cryptography is the computation of the pairing function, which has been known [to be] even more time-consuming than the exponentiation function in RSA. Fortunately, fast hardware implementations of the pairing have been reported recently. ... For example, it is reported ... that the Tate pairing can be calculated in about 6 ms on a [field-programmable gate array (FPGA)].

When such notions exist, a reader may want to know how efficient the authors’ implementation is, compared to other architectures. The lack of diagrams showing the actual details of the proposal is also a drawback, because some readers may need this information to follow the authors’ line of thinking. The paper does not include results or comparisons with other existing architectures. In conclusion, I cannot recommend this paper, even though the idea it presents does show some promise.