##Reference
Reviewer 1
The paper does not discuss, with enough detail, some of the related work.
Specific topics such as bandwidth prediction for flows and PBAC/MBAC are fairly (albeit not extensively) covered. However, the core question of this work (“how to improve QoS for unmanaged LANs or home networks?”) is not properly covered from a state-of-the-art perspective. In order to rationalize their proposal, authors should compare it with the alternative approaches that also address this goal. I understand that by focusing their proposal on bandwidth reservations authors already assume a distinction from a substantial part of previous work, but I would like to see a broader analysis of this field - otherwise the reader may raise doubts about the merit of the selected approach.
From the industry side I would point:
– ITU-T H.622 framework (event though H.622 assumes some level of operator-assisted management, unlike the proposed framework).
– UPnP-QoS, which is also extensively studied in academic literature.
From a more academic point of view, I would point for instance these papers:
– Suraci et al., “Admission Control and Drop strategies in a UPnP-QoS controlled home network”, 10.1109/PIMRC.2010.5671830
– Castrucci et al., “Application QoS management and session control in a heterogeneous home network using Inter-MAC layer support”, http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5722465&isnumber=5722317
– Chen et al., “QoS management in heterogeneous home networks”. Computer Networks, 2007 (despite its focus on DiffServ).
Many references lack the title of the paper, including only the title of the book or journal the paper was published in. This is disturbing and should be corrected.