Six Challenges in Home Network Configuration, Control and Troubleshooting
Summary:
We are interested in funding efforts that focus on one or more of six
practical problem areas for today's home networks. These problems are the
Failure of Plug and Play, Configuration Complexity, the need for Alternative
Configuration Paradigms, the lack of effective Access Controls, issues of
Public Versus Private Networking, and Centralized Versus Distributed Control of home network services.
Full Description:
There is a wide gulf between what users can do on their home networks today
and what future applications demand. New applications such as in-home media
and energy management need a new level of reliability and serviceability
from home networks. Today, however, network configuration and control
problems are widespread on self-managed - and unmanaged - home gateways,
network-attached storage, media servers, and other products. There are
several configuration and control problems in the current state of the art
that may benefit from more research attention.
The Failure of Plug and Play: Over ten years ago, the UPnP Forum and
others set out to make home networks "plug and play", using an analogy from
PC peripheral installation. Unfortunately, while UPnP IGD (Internet Gateway
Device protocol) is widely shipped, it is not widely used due to a dearth of
management applications (i.e. IGD control points). The entire industry lacks
methods for simple configuration and control of residential gateways and
other network devices. Instead, automated tools for configuring gateway
services such as DNS addresses, NATs and firewalls have led to some
well-publicized attacks.
Configuration Complexity: Easy setup of consumer network equipment
relies on PC installation scripts, which have a "day zero" vulnerability in
their use of a well-known password. Although installation scripts shipped on
CDs, DVDs, USB drives and other media may always be needed by some users,
others don't or can't use this method, e.g. when there is no PC interface,
such as on handheld devices, or when an operating system can't run the
installation script. If the installation CD isn't used, or some setting
needs to be reset, people must use system configuration screens. The current
practice of giving total system configuration access to the user is
confusing to many. Giving programmatic tools (like UPnP) total access gives
the tool too much privilege. Alternative methods need a different paradigm
such as social networking or outsourcing of configuration and management.
Alternative Configuration Paradigms: Alternative paradigms for the
configuration and control of home networks have had some success to date.
The WiFi Alliance has developed 'mental models' for device introduction that
include pushing buttons, entering a number or other actions that are
specific to the purpose, application or device. Although products have
shipped with such features, most product developers and vendors today
persist in having users configure products using a PC installation CD or by
editing the system configuration in a GUI.
Access Controls: Home gateway/routers, network attached storage,
servers and other shared, network devices have rudimentary access controls
that ignore established security principles such as least privilege. The
lack of suitable access controls enables malware and other interlopers to
subvert critical network services. Tasks like configuring shares on network
storage or granting privileges to utility providers' devices for energy
management are beyond the capability of today's typical home network user.
Public Versus Private Networks: Earlier LAN protocols such as
NetBIOS and AppleTalk have evolved into today's UPnP, Bonjour, and other
"private" or "local scope" protocols. In the near future, dual-stack
IPv4/IPv6 home networks offer private and public addressing for Bonjour,
UPnP, sensor networks, and other services with private scopes such as Unique
Local Addressing and Link Local Addressing. There is little guidance
available, however, for using scoped addressing on home networks that
include mobile as well as fixed-location devices and a more-than-sufficientsupply of global IPv6 addresses.
Centralized versus Distributed Control: Home network protocol suites
such as UPnP and Bonjour use distributed algorithms for naming, addressing,
and service discovery on dual-stack Internet Protocol networks. Highly
distributed solutions have known problems, however, such as in revoking a
device that has been lost or stolen. Centralized solutions such as
outsourced management solve some of these problems. We are interested in the
tradeoffs of centralized versus distributed management solutions.
We are looking for novel approaches to these and related problems. Cisco is
interested in relevant work by researchers in computer science and
engineering fields such as networking, security and in interdisciplinary
areas such as HCI and home automation.
Constraints and other information:
Please use the link below to submit a proposal for research responding to this RFP. After a preliminary review, we may ask you to revise and resubmit your proposal.
Proposal submission:
Please use the link below to submit a proposal for research responding to this RFP. After a preliminary review, we may ask you to revise and resubmit your proposal.
RFPs may be withdrawn as research proposals are funded, or interest in the specific topic is satisfied. Researchers should plan to submit their proposals as soon as possible. The deadline for Submissions is the Friday of the first week of each calendar quarter (the months of January, April, July, October). Funding decisions and communication will occur within 90 days from the quarterly submission deadline. The usage of funding is expected within 12 months of funding decision. Please plan your requests accordingly.
