1. Rating
  Technical quality
    3: Some errors which can be easily corrected, as described below.
	There are some conflicts, where I don't understand. For example, 
	the paper states "every pair of nodes cannot reach each other within 
	one hop", but later, it takes the "cliques" in the assumption, which 
	indicates the one hop distance within the cluster. Further, if all 
	participants stays in the radio range. There is no need to relay at all.
    
  Originality
    2: Good survey of current state; no new ideas

  Editorial quality
    3: Understandable with some effort, several improvements suggested
       below
       Since the paper described a lot on key establishment and different 
       hierarchies and clustering topologies, diagrams and comparisons are 
       suggested to provide for a clear view. The procedure to generate a 
       group key and other temporary keys (e.g. blind key, session key) could 
       be illustrate with a demostrative pictures. At last, the security 
       requirements were stated, but there is no analysis of the algorithm 
       about how it meets the requirements afterwards.

  Overall grade (overall, how do you rate this paper?)
    2.5: Good
  
  Confidence (how confident are you about this review?)
    2: I have some general knowledge of this subject

2. Detailed comments
  2a. Technical quality
The paper introduced a new solution for clustered AKE. It clearly addressed
several key concepts, e.g. cluster, cliques and various tree topologies. But
it fails to convince me why this new solution is superior to other approaches.
Also, it does not describe existing schemes and ad hoc particularity well. 
Section 3 and 5 need more effort.

  2b. Originality
The paper does not propose really new ideas, but rather a technique
description and review. It could be reinforced by giving a deep 
analysis of the described solution, for example, how efficient it is?
any use cases or scenarios? Comparison of the complexity with other
solutions.

  2c. Editorial quality
The language is ok, but more presentation manners are strongly suggested to
provide while describing the key establishment process and structure. Diagrams,
table and list make it much clearer.


Good references:

References seems OK. Nothing to say.