======= PIMRC 2017 Review 1 ======= > *** Relevance and Timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Good. (4) > *** Technical Content and Scientific Rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Solid work of notable importance. (4) > *** Novelty and Originality: Rate the novelty and originality of the ideas or results presented in the paper. Significant original work and novel results. (4) > *** Quality of Presentation: Rate the paper organization, the quality of text, English, and figures and the completeness and accuracy of references. Well written. (4) > *** Strong Aspects: Comments to the author: What are the strong aspects of the paper? This paper studies EPC placement in LTE networks to establish local deployments. The topic is very relevant and up to date for example in the public safety domain. The paper provides useful contributions to the topic in terms of studying where the local EPC should be located within the network. > *** Weak Aspects: Comments to the author: What are the weak aspects of the paper? The paper specifically focuses on the flow optimization in order to determine the location for the EPC. There is no proper description on the big topic of local EPC including the tasks of the local EPC in comparison to the traditional EPC. As it is a new approach, it should be thoroughly motivated and explained which would be interesting for the readers. > *** Recommended Changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. Please, provide a complete description of the local EPC approach before jumping into the details of optimal placement in terms of the traffic flows. > *** Submission Policy: Does the paper list the same author(s), title and abstract (minor wording differences in the abstract are ok) in its PDF file and EDAS registration? Order of authors is different in pdf and EDAS registration. ======= PIMRC 2017 Review 2 ======= > *** Relevance and Timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Good. (4) > *** Technical Content and Scientific Rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Solid work of notable importance. (4) > *** Novelty and Originality: Rate the novelty and originality of the ideas or results presented in the paper. Some interesting ideas and results on a subject well investigated. (3) > *** Quality of Presentation: Rate the paper organization, the quality of text, English, and figures and the completeness and accuracy of references. Well written. (4) > *** Strong Aspects: Comments to the author: What are the strong aspects of the paper? Well structured Good topic > *** Weak Aspects: Comments to the author: What are the weak aspects of the paper? Not enough critical discussion > *** Recommended Changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. More critical discussion > *** Submission Policy: Does the paper list the same author(s), title and abstract (minor wording differences in the abstract are ok) in its PDF file and EDAS registration? N/A ======= PIMRC 2017 Review 3 ======= > *** Relevance and Timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Good. (4) > *** Technical Content and Scientific Rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Valid work but limited contribution. (3) > *** Novelty and Originality: Rate the novelty and originality of the ideas or results presented in the paper. Some interesting ideas and results on a subject well investigated. (3) > *** Quality of Presentation: Rate the paper organization, the quality of text, English, and figures and the completeness and accuracy of references. Well written. (4) > *** Strong Aspects: Comments to the author: What are the strong aspects of the paper? The paper proposes a novel metric of access nodes and the corresponding network function placement optimization focusing on the flow capacity characteristic, which realizes the maximization of the traffic amount in the mobile network with isolated base stations. The paper is written in a generally well-organized structure and the result analysis is demonstrated with various graphs in an intuitive and lucid manner. > *** Weak Aspects: Comments to the author: What are the weak aspects of the paper? In the performance analysis section, the proposed metric is compared with previous ones from one single aspect in term of traffic amount, which may cause a lack of comprehensiveness. Besides, the title of the paper is too general and vague, thus not able to specify the core feature of the work. > *** Recommended Changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. Different aspects of the performance should be analyzed together in a more thorough fashion in the verification section. The title should also be modified to a more detailed one that can accurately convey the contribution of the work. > *** Submission Policy: Does the paper list the same author(s), title and abstract (minor wording differences in the abstract are ok) in its PDF file and EDAS registration? yes ======= PIMRC 2017 Review 4 ======= > *** Relevance and Timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Good. (4) > *** Technical Content and Scientific Rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Valid work but limited contribution. (3) > *** Novelty and Originality: Rate the novelty and originality of the ideas or results presented in the paper. Significant original work and novel results. (4) > *** Quality of Presentation: Rate the paper organization, the quality of text, English, and figures and the completeness and accuracy of references. Well written. (4) > *** Strong Aspects: Comments to the author: What are the strong aspects of the paper? This paper proposes a new scheme for solving the local EPC placement problem in the network in order to determine which base stations the Local EPC must be co-located. In addition a novel centrality metric is proposed for measuring the capacity of a node to receive the total amount of flows in the network. > *** Weak Aspects: Comments to the author: What are the weak aspects of the paper? The entrality metric is proposed for measuring the capacity of a node to receive the total amount of flows in the network would need elaboration to include realistic requirement is terms of varying BS demands and end-to- end characteristics. constraints must be further included in the Local EPC placement criteria. > *** Recommended Changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. The authors are kindly requested to elaborate the proposed proposed for measuring the capacity of a node in support of end-to-end delay characteristics and constraints. > *** Submission Policy: Does the paper list the same author(s), title and abstract (minor wording differences in the abstract are ok) in its PDF file and EDAS registration? OK #################################################################################################################### ########################### Reviewer comments for previous version submitted at ICCCN 2017 ####################### #################################################################################################################### ----------------------- REVIEW 1 --------------------- PAPER: 139 TITLE: Core Network Function Placement in Mobile Networks AUTHORS: Jad Oueis, Vania Conan, Damien Lavaux, Razvan Stanica and Fabrice Valois Overall evaluation: 1 (weak accept) ----------- Overall evaluation ----------- This paper proposes flow centrality which measures a node capacity of receiving the total amount of flows in the network. The authors show the advantages by comparing the flow centrality to other centrality metrics. Overall, the paper is well organized, from problem statement, to solution and finally to simulation. But in the capacity distribution, the authors consider the probabilistic model, the distribution of capacity can be unfair. For example, assuming that there are 10 nodes, the maximum capacity is 100, but probabilistically all nodes can have a capacity 10. Therefore, it is need to know that in case of unfairness. Also the authors considered the number of nodes to 10 in figure 4-5. But it is necessary to consider the case where a large number of nodes. ----------------------- REVIEW 2 --------------------- PAPER: 139 TITLE: Core Network Function Placement in Mobile Networks AUTHORS: Jad Oueis, Vania Conan, Damien Lavaux, Razvan Stanica and Fabrice Valois Overall evaluation: 1 (weak accept) ----------- Overall evaluation ----------- Authors of this paper investigated the Local EPC placement problem. The main contribution of this paper is the flow centrality metric, which is developed to measure the nodes’ capacity of receiving the total flow in a network. I have two comments on this paper. 1) As formulated in Eq. (2), the main objective is to maximize \lambda(d), but \lambda(d) is defined in Section IV-A as the traffic that all eNodeBs in the network send towards the Local EPC. The question is how can we maximize the total traffic generated by eNodeBs. I guess you intend to place a Local EPC to the location where the node (d) has the maximum traffic, right? Authors should clarify this point and provide detailed explanations on it. 2) It is not clear how to incorporate the proposed metric to the author's solution. ----------------------- REVIEW 3 --------------------- PAPER: 139 TITLE: Core Network Function Placement in Mobile Networks AUTHORS: Jad Oueis, Vania Conan, Damien Lavaux, Razvan Stanica and Fabrice Valois Overall evaluation: 1 (weak accept) ----------- Overall evaluation ----------- In this paper, The auuthors attempt to solve a placement problem of a Local EPC serving a network. The authors defined new metrics such as flow centrality the flow centrality. The authors then prove that the proposed approach can achieve optimal flow centrality. In addition, the authors compared the flow centrality to other state of the art centrality measures, and highlighted the loss. The research is well motivated. Some ideas are novel ----------------------- REVIEW 4 --------------------- PAPER: 139 TITLE: Core Network Function Placement in Mobile Networks AUTHORS: Jad Oueis, Vania Conan, Damien Lavaux, Razvan Stanica and Fabrice Valois Overall evaluation: -1 (weak reject) ----------- Overall evaluation ----------- The authors propose a new metric: "flow centrality" to optimize placement of local EPC in LTE networks. It is unclear how the authors solve the optimization problem in eq. (2) and whether this problem is hard. The paper also lack proper analysis of the proposed formulation. The evaluation is weak and unconvincing. This new metric is evaluated on Grid graphs and Random geometric graphs. Hence, It is not clear that the topologies on which the authors evaluated their approach are realistic or match real LTE networks. The parameters of the evaluation are also ad-hoc. It is not clear how the authors choose them why they make sense in real LTE networks. #################################################################################################################### ########################### Reviewer comments for previous version submitted at ICC 2017 ######################### #################################################################################################################### ======= Review 1 ======= > *** Submission Policy: Does the paper list the same author(s), title and abstract in its PDF file and EDAS registration? They are all the same. There is just a minor difference, probably negligible, in the details of affiliation for J. Oueis, R. Stanica and D. Lavaux. > *** Strong aspects: Comments to the author: what are the strong aspects of the paper Summary. The authors deal with the problem of choosing the best location for a single backhaul to the traditional core network, in an LTE network. In more details: given a network of nodes (eNodeB), connected by links with different bandwidth, the goal is to find the node location for the link to the Internet (Local EPC), so to maximize the global throughput with the assumption that all the nodes generate the same amount of traffic. The authors propose a novel centrality metric, flow centrality, that measures a node capacity of receiving the total amount of flows in the network. The choice of the optimal node location is then given by the solution of an optimization problem. Strong aspects. The proposed flow centrality metric is novel, taking into account the weights of all the links of the network. The authors compare their solution with other solutions from literature showing the effectiveness of their solution. > *** Weak aspects: Comments to the author: what are the weak aspects of the paper? The authors say that their solution requires a few minutes of computation to resolve the optimization problem, in the example case proposed in Fig. 2a. It's not clear how this time increases, when increasing the size of the network. In other words, it’s not clear what is the complexity of the optimization problem. > *** Recommended changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. The authors could add some consideration about the complexity of the optimization problem to resolve in eq. (5). > *** Relevance and timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Good (4) > *** Technical content and scientific rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Solid work of notable importance. (4) > *** Novelty and originality: Rate the novelty and originality of the ideas or results presented in the paper. Significant original work and novel results. (4) > *** Quality of presentation: Rate the paper organization, the clearness of text and figures, the completeness and accuracy of references. Well written. (4) ======= Review 2 ======= > *** Submission Policy: Does the paper list the same author(s), title and abstract in its PDF file and EDAS registration? Yes. > *** Strong aspects: Comments to the author: what are the strong aspects of the paper The focus of this paper is on EPC placement, with different constraints on flow centrality. The authors propose an optimization framework to address this problem. This problem is of importance in the design of modern backhaul systems and is hence useful. > *** Weak aspects: Comments to the author: what are the weak aspects of the paper? I would recommend addition of more intuition into why the proposed algorithm works. As of now, it just seems like a black box type approach. I would also like to see more performance studies and benchmarking with different EPC design algorithms in the literature. I do not see many comparisons with the literature. > *** Recommended changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. Address the weak points. In particular, address the intuition and better comparison aspects. > *** Relevance and timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Acceptable (3) > *** Technical content and scientific rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Valid work but limited contribution. (3) > *** Novelty and originality: Rate the novelty and originality of the ideas or results presented in the paper. Some interesting ideas and results on a subject well investigated. (3) > *** Quality of presentation: Rate the paper organization, the clearness of text and figures, the completeness and accuracy of references. Readable, but revision is needed in some parts. (3) ======= Review 3 ======= > *** Submission Policy: Does the paper list the same author(s), title and abstract in its PDF file and EDAS registration? Some authors' affiliations in the manuscript is a little different from those registered to the EDAS system. > *** Strong aspects: Comments to the author: what are the strong aspects of the paper - Novel algorithm for placing the Local EPC among eNodeBs is proposed. - Performance of the proposed algorithm is compared with other methods through some simulation experiments. > *** Weak aspects: Comments to the author: what are the weak aspects of the paper? - A little difficult to follow due to the lack and poor legibility of figures. - The versatility of the proposed algorithm should be discussed. > *** Recommended changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. The proposed algorithm based on the flow centrality seems to be effective in the environment the authors assume. The reviewer supposes, however, that the problem setting is similar to the server placement problem in wireless mesh networks and the sink placement problem in wireless sensor networks, and that the proposed approach can be applied to such other environments. If there are any specific constraints in the assumed Local EPC placement problem, the authors should discuss them. If not, the authors should discuss the versatility of the proposed approach. In addition, all figures are too small. This makes the legibility of figures quite low. The authors should enlarge them in order to improve the legibility. It is questionable whether the relative loss is appropriate metric for comparing the performances of the proposed and other approaches. The reviewer supposes it is better to show some absolute values of each approach and compare those values. For example, it is reasonable to evaluate the throughput changing the load from each nodes. > *** Relevance and timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Good (4) > *** Technical content and scientific rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Valid work but limited contribution. (3) > *** Novelty and originality: Rate the novelty and originality of the ideas or results presented in the paper. Significant original work and novel results. (4) > *** Quality of presentation: Rate the paper organization, the clearness of text and figures, the completeness and accuracy of references. Substantial revision work is needed. (2) ======= Review 4 ======= > *** Submission Policy: Does the paper list the same author(s), title and abstract in its PDF file and EDAS registration? yes > *** Strong aspects: Comments to the author: what are the strong aspects of the paper The authors study the relevant problem of placing core network (i.e. EPC) functions at the network edge. To do this, the authors select a new metric called flow centrality. eNodeBs are assumed having no backhaul to the traditional EPC, which renders the general setup interesting to study. The paper is generally well written, easy to follow and figures are self-explanatory. Results are interesting and sound. > *** Weak aspects: Comments to the author: what are the weak aspects of the paper? Though the paper is generally well written, the abstract can be improved: “…must be co-sited with in a network”, "within" seems more appropriate for example. As in any multi-hop network, the eNodeB with which the EPC is collocated will be the bottleneck. This aspect is not at all discussed in the paper. It is true that the node with the maximum flow centrality will probably be that with lower bottleneck problems, which anyway do not disappear entirely. This reviewer suggests to make this point clearer. What technology is used for inter-eNodeB connectivity? Here there will be another bottleneck. This reviewer suggests to make one or two examples. To this reviewer it is not completely clear the statement made in the left hand column of the introduction “…This would allow the Local EPC to receive a maximum amount of traffic from all the base stations, and vice versa”. Traffic has to go anyway through the local EPC, there is no way out otherwise. The fact that the EPC is co-sited with the eNodeB with maximum flow centrality looks like to implicitly relieve possible bottleneck problems. Is this what the authors claim, though not explicitly said with these words? The impression of this reviewer is that the authors do not completely underpin the advantage provided by the flow centrality metric. > *** Recommended changes: Recommended changes. Please indicate any changes that should be made to the paper if accepted. Please, refer to the weak aspects and incorporates the necessary changes in the paper. > *** Relevance and timeliness: Rate the importance and timeliness of the topic addressed in the paper within its area of research. Good (4) > *** Technical content and scientific rigour: Rate the technical content of the paper (e.g.: completeness of the analysis or simulation study, thoroughness of the treatise, accuracy of the models, etc.), its soundness and scientific rigour. Valid work but limited contribution. (3) > *** Novelty and originality: Rate the novelty and originality of the ideas or results presented in the paper. Significant original work and novel results. (4) > *** Quality of presentation: Rate the paper organization, the clearness of text and figures, the completeness and accuracy of references. Well written. (4)