|
Securing the wireless LAN
Wi-Fi
and wireless LANs are the rage today, with most new notebooks and PDAs having
built-in features to access WLANs. However, despite the eagerness of users,
security remains a huge concern when it comes to WLANs. Dhananjay Ganjoo
describes the security issues that WLANs face and the solutions
True mobility does not come easy. Not if it has to be combined with high security
and seamless roaming. Making sure that you provide uninterrupted service as
you securely roam across 802.11a and b, multiple access points, and subnet boundaries
is a daunting task by itself, but add a huge dose of high security to it and
you realise the complexities go deeper.
First generation wireless LAN (WLAN ) systems were all about basic standards,
connectivity and end-user benefits. But the entry of second generation WLAN
systems has brought about a strategic change and the onus is on enhanced standards,
addressing security, quality of service (QoS) and interoperability in enterprise-wide
roaming. And, of course the undergrid is all about optimal pricing, performance
and control.
And today security is way on top of these requirements.
True mobility comes at a price
Modern business requires mobility and always-on connectivity. Hence, WLANs,
now a standard feature in many PDAs and notebooks, are being deployed widely,
particularly in hospitals and universities, in homes, and in thousands of hotspots
in coffee shops, hotels, and airports. Even India is not immune to the WLAN
revolution and more and more hotspots are coming up everyday.
But the freedom mobility brings comes with a price: wireless networks do not
respect the boundaries of solid walls or locked doors; movement sensors and
security cameras cannot detect unwanted guests on your network. Unchecked, this
kind of unrestricted access leaves corporations wide open to attack and theft
of corporate data.
Today, organisations are looking at a security strategy that encompasses wireless
even if they dont have a plan to deploy wireless; but free agents
or rogue access points could exist in the enterprise and effectively
put the network at risk.
A free agent is a wireless device that has been connected to the
network without the permission or knowledge of the IT department. A well-meaning
employee, who well call Bob has some knowledge of IT and goes
down to his local computer store and buys a cheap wireless solution. Bob plugs
the wireless device in under his desk. While Bob did not do this for nefarious
purposes, he has exposed the corporation to a security risk. The radio signal
from Bobs access point does not stop at the walls to his office. It extends
out into the street where someone could easily sniff out the traffic and potentially
break into the network.
A rogue access point is one that has been installed specifically
for nefarious purposes inside or outside the office. The rogue access
point pretends to be part of the corporate network with the intention of intercepting
traffic such as authentication information from wireless users. Armed with this
information, someone could potentially break into the network.
Gartner Group estimates that there are over 10 million WLAN users globally,
with that number doubling in 2004 and a Gartner study suggested that enterprises
could expect a 22 percent productivity improvement. But going by the actual
deployment and faith reposed in WLAN, one realises the main hindering factor
is security.
Behind the scenes
Enterprises are increasingly realising that by adding a wireless node to the
corporate network they must also include appropriate security precautions and
good security practices.
A typical ratio of WLAN clients to access points is 10:1, although this varies
between technologies and vendors. Overlapping access point cells of coverage
are created when multiple access points exist around the floor or building,
creating a single coverage area for mobile users.
In the first generation of WLANs, the world of wireless data networking was
dominated by proprietary niche products that targeted specific vertical markets.
The development and wide vendor acceptance of WLAN IEEE 802.11 standards has
changed this, and resulted in increased availability of PC and PDA interface
cards and plummeting prices.
Standards such as 802.11a and b delivered speed and connectivity, and end-users
loved it. However, the standards did not solve everythingand in fact,
the first standards were found to have many shortcomings. Worse, the deployment
of access points and security mechanisms was done in an ad hoc fashion, leaving
IT departments with issues on a number of fronts.
First generation challenges
The security exposures of using WLANs have been well documented, including identifying
non-secure access points (APs) by war-driving and war-chalking,
the inadvertent insertion of free agent access points and the malicious insertion
of rogue access points.
Wired Equivalent Privacy (WEP), the primary security mechanism that shipped
with most WLAN products, has also proven to be non-secure and opens up the network
to unauthorised access, session hijacking, eavesdropping, and other threats.
Limited secure mobility (WLAN users cannot generally move between subnets without
re-authenticating themselves with the network) and lack of access and bandwidth
controls, and no QoS for multimedia applications and IP telephony are the other
hurdles.
WLANs are becoming increasingly difficult to managea problem exacerbated
by the lack of adequate security, access, and bandwidth controls. For example,
if there is no ability to control users from hogging wireless bandwidth, overall
WLAN performance is impacted and audit trails are non-existent. Sometimes, even
knowing where the access points are physically located is a challenge.
An outsider is blocked from seeing the internal WLAN because the outside APs
operate at the same carrier frequency as the internal ones and offer a higher
signal strength to the outsider, thereby in effect, jamming the
internal signal for the outsider. The disadvantage of this approach is that
it is not only expensive, but it cannot be 100 percent effective.
Second generation WLAN
Second generation WLAN is an attempt at creating a secure architecture to put
IT back in control. The foundations of second generation WLAN solutions that
meet the needs of enterprises consist of standards. The IEEE 802.11 committee
has responded to the needs of second generation WLAN users by undertaking the
development of a number of new standards, which complement IEEE 802.11a and
11b. Most notable among these is 802.11i, which establishes robust WLAN infrastructure
for security.
Other standards being finalised address WLAN QoS to allow IP telephony and multimedia
application support, and multi-vendor interoperability of roaming handovers
across access points.
Second generation challenges
All these standards present a challenge of determining which standard to use
when and where. Ideally, the 802.11 device would be able to sense what the AP
is running (802.11a, b, g, d, and/or h) and be able to dynamically switch to
connect to it. The AP may run both 802.11a and b simultaneously, or be configurable
to 2.4 or 5 GHz depending on the desired coverage area or local interference.
5 GHz has more attenuation when going through walls and windows. In the end,
devices should be able to roam seamlessly from one standard to another (e.g.,
802.11a - b).
Layered architecture is the solution
Enterprises need one standards-based secure WLAN architecture that supports
WLANs as an inherent manageable and secure part of their infrastructure. The
requirements of such an architecture must address the issues associated with
first generation WLAN systems, recognising that a layered approach to WLAN functionality,
including security, is required to meet the varying needs of enterprises.
WLAN architecture, based on a layered approach both physically and functionally
will allow the optimal distribution of functionality and security for performance
and low total cost of ownership.
Access points compose the first layer of second generation WLAN architecture,
providing wireless connectivity to roaming mobile users equipped with notebooks,
PDAs, and telephones. These access points are designed to evolve to support
new wireless standards and technologies that allow more effective use of the
radio spectrum and provide security over the radio link. Adding functionality
to access points above the first or second layer (even if feasible) has a significant
impact on TCO, because of the highly distributed nature of AP deployment. In
addition, certain functions, such as enterprise-wide roaming (and ultimately
seamless roaming across enterprise and public domains), can be better handled
in more centralised devices that support multiple APs.
The second layer of this architecture is wired Ethernet networking, which has
support for Power over Ethernet, VLAN segmentation and QoS capabilities. APs
are connected to Ethernet switches, which provide Power Over Ethernet (using
the draft standard IEEE 802.3af). These Ethernet switches are either dedicated
to WLAN aggregation or are shared with the wired LAN network with segregation
provided via virtual LANs.
The advantage lies in the fact that the enterprise has the choice of where and
how it wants to integrate WLANs into the basic wired Ethernet technology. It
also allows a common Power over Ethernet technology to be used consistently
over wired and wireless environments.
The third layer provides networking and application-aware security at Layer
3 to 7 of the OSI model. And the WLAN security switch interfaces to the enterprise
and to policy management , including directories and policy servers, which constitute
the fourth layer of the architecture.
This four-layered WLAN architecture provides a high degree of flexibility while
meeting the needs of the enterprise for secure WLAN access. It is complemented
by access control, which authenticates all users and authorises which network
resources are accessible for both the wireless and wired portions of the network.
The author is national sales manager, Nortel Networks India. He can be contacted
at dganjoo@nortelnetworks.com
| Yesterdays environment
First generation WLAN systems
were all about basic connectivity standards and end-user benefits, much
the way Ethernet in its early days evolved around ad hoc networking, sharing
network resources, and unstructured wiring. This is also how the Internet
started. This represents yesterdays environment and todays
opportunity.
Second generation WLAN systems
are all about enhanced standards, addressing security, QoS and interoperability,
and architected solutions focusing on optimal price, performance, and
control. IP mobility will open the door for roaming across the enterprise,
not just across a few wireless cells. This phase is quite analogous to
the widespread adoption of in-building Layer 2 to 7 architectures based
on switched Ethernet and hierarchical campus networks built around routing
switches.
Tomorrows dream
Third generation WLAN systems
are all about bringing down the boundaries between enterprise WLAN systems
and public wireless systems for seamless roaming, extending the application
of IP mobility standards, and comprehensive security and signalling mechanisms.
Next-generation (so-called 2.5 and 3G) public wireless systems and a plethora
of new mobility devices will have a dramatic impact on the way business
works, delivering up to 1.5 Mbps throughput for data.
|
|