Wireless TechTalk Category Archives

March 28, 2008

Architecting a Reliable Wireless Network

Few question that the age of mobility is upon us. The emergence of a global economy is facilitated by the basic mobility of goods, services and people. For many, the idea of mobility has become second nature. Of course, the structures for mobility – transportation, logistics, infrastructure – must be robust and reliable lest the mobile quickly become immobile.

Given this, it is critical that the wireless industry focus R&D on improving network performance and reliability. What makes this challenging is the need to do this from applications, to the network, and very importantly, to a wide range of mobile devices. To achieve this, businesses should focus on the following key areas when designing a highly available wireless network:

• Managing the growth in wireless devices – Managing the growth and diversity of devices connecting to the network requires a continued focus on device compatibility and testing. Through its Cisco Compatible Extensions program, Cisco works with leading Wi-Fi silicon manufacturers to ensure simple, secure connections for a broad range of mobile devices. Another key need is ongoing interoperability testing and validation. While not widely known, Cisco has a very robust program in place, called Assurewave, to achieve just this.

• RF management is critical – IT should not underestimate the importance of pushing the envelope in radio development and optimization. In the wireless industry, providers either integrate off-the-shelf radio technology, or internally develop radios. Only by internally developing its own radios and antennas can a provider truly deliver a high performance access point. A robust solution combines radio expertise with software excellence so IT can benefit from dynamic radio resource management capabilities. This removes the complexity of management while ensuring an adaptive network capable of responding to environmental changes including changes in user and device location and density.

• Spectrum Intelligence – It is difficult to place heavy reliability requirements on wireless networks without clear visibility into the RF spectrum. Comprehensive spectrum intelligence solutions are critical to provide the intelligence to classify, locate and mitigate RF interference, including from non Wi-Fi sources such as microwave ovens, cordless phones and Bluetooth.

• System Unification – For performance and simplicity, the wireless network must be unified with the wired network. In this way, IT can leverage common services and provide the processing horsepower needed to support the breadth of mobility applications being demanded by users, while achieving efficiencies in power and resource consumption.

• Intuitive Management – By deploying a comprehensive management solution, IT can obtain the requisite diagnostics and reports on system health before issues become service impacting. The wireless management solution must have proactive tools that help IT optimize the wireless deployment and on going operations.

Only by deploying a wireless infrastructure designed to meet the reliability needs of the most stringent environments can IT enable the business to take advantage of emerging mobility solutions.

Cisco has led the industry in the development of industry specific solutions like mobile care solutions for healthcare and in-store mobility solutions for retail. These are delivered over next generation, high performance wireless networks like 802.11n. With a significant part of its R&D invested in developing highly available wireless networks, Cisco continues to turn the immobile into the mobile, and keep them there.

Posted by Chris Kozup at 01:00 PM Permalink | Comments (0) | TrackBacks (0)

November 05, 2007

How to defeat the laws of physics

In my blog posting on enterprise WLAN, I talked about how the 2.4 spectrum simply does not have enough capacity for large scale deployments. My thesis was that users should really be looking to maximize the 5Ghz spectrum for their use (while recognizing that legacy devices will continue to exist in the 2.4 spectrum). Following this article, some vendors argued that existing WLAN architectures could not meet the needs of the enterprise, and the only valid architecture was one that used the same channel across all APs. They claim that a single channel architecture is the only one that can support fast roaming and provides maximum performance. In this blog series, I will explain why this is myth.

Vendors of these single channel architectures claim that handoffs are too slow and cannot support the needs of real-time applications. I think it is important that we be intellectually honest when we discuss roaming speeds. First, I admit that in the old world of stand-alone APs, where no centralized coordination existed to help improve roaming times, this was an issue. There are two main functions that a client needs to do when it decides it will roam. First, it needs to find an alternative access point and then needs to provide its credentials to gain access to the network.

Finding a candidate access point typically requires a client to scan all of the available channels, and on a dual mode (2.4Ghz and 5Ghz device), this is quite a number of channels. This process can be sped up through some innovation on the client side, but is best when it is assisted by the infrastructure. Cisco supports the neighbor discovery extension, which allows an access point to transmit its neighboring APs, along with their channels. This information can be used by clients to minimize the number of channels to scan.

Once the client has found a new access point, it needs to request access, which includes the authentication process. The issue with re-authentication is the fact that it requires that the backend RADIUS server be involved, which increases latency, and the overall cost of the handoff. Eliminating the authentication process altogether, and only relying on a series of transactions between the client and the access point, is ideal. To do this, Cisco uses CCKM to allow a device to simply re-key, using information that was distributed during the original authentication phase. CCKM is supported on both the autonomous and unified wireless architectures, and the latter can support roaming times of sub-20ms.

Cisco is working with the IEEE 802.11 to create an industry standards. Neighbor discovery will be available in 802.11k, while 802.11r will include CCKM- like features. This standard involves many of the consumer and enterprise class access point vendors, all of which are working diligently to provide these technologies to their customers quickly. Given that we've already proven that handoffs times can support the needs of real-time applications, I can only assume that the argument being made by these vendors is against access point technologies as they existed five years ago.

Next time, we will focus on RF related issues brought up by these vendors.

Posted by Pat Calhoun at 07:15 AM Permalink | Comments (0) | TrackBacks (0)

October 04, 2007

What's up with MIMO? - Part Deux

In my first episode on MIMO (What's up with MIMO), I introduced the concept of multiple antennas and the home audio analogy. In this example, I discussed one particular mode of MIMO that transmits the same stream across all antennas, and makes use of multipath to increase the chance of correctly decoding the received signal - thus lowering the bit error rate. This method exploits what is known as spatial diversity, which is one of the simplest method to achieve MIMO gains. But MIMO is much more than that.

There is also another mode of MIMO that is ideal in environments that experience few errors, which is called spatial multiplexing. In this mode, the transmitter would treat each antenna as a "separate channel" - much like you would experience in a 5 channel home audio system. These systems send the audio stream that provides directionality to the whole experience. Of course, in MIMO systems, we don't care much about directionality, but we certainly care about sending multiple streams to increase the overall effective bandwidth.

So for instance, if an AP wanted to send a packet, it could split the packet up into three "chunks", and transmit each "chunk" across each of its three antennas simultaneously. This would effectively reduce the transmit time - and therefore increase the overall capacity of the network.

Posted by Pat Calhoun at 11:26 AM Permalink | Comments (4) | TrackBacks (0)

September 17, 2007

Your Network's Worst Enemy

Wireless networks have matured to the point where security is no longer their primary concern. More and more I am hearing from customers that their primary issue these days is interference.

Interference is one of those things that is really tricky to both identify and resolve. There are few IT geeks out there that I've met, who happen to understand networking and the complexity of radio frequencies. In fact, I wouldn't be surprised if most people reading this blog entry have never even seen a spectrum analyzer. I know I have, but it has to be a very bad day when I need to turn one on... It is in fact for this main reason that as we designed the unified Wireless LAN controller - to take the science out of RF and try to provide more visibility in a manner that most networking experts understand. We turned that complex spectrum analyzer into a graph that would show the health of the network, by showing both interference (other 802.11 traffic) and noise (non 802.11 traffic). We've heard consistently from our customers how much they appreciate getting a better understanding of their network.

The past year has been a very exciting time for us in the Wireless LAN business, as we've seen the technology grow from a convenience to a mission critical network. Ironically, while I think we all agree of the importance of this network, most IT managers would still prefer to treat it as a secondary network - and I believe that this is mostly because the science behind RF makes it a very difficult network to troubleshoot. What does one do when excessive noise shows up? It's not quite like calling your neighbor and asking them to turn down the Deep Purple they've been blasting since 9PM (ok... I'm showing my age... these days you'd probably be complaining about Linkin Park). The expertize, and equipment, required to identify the source of noise is so complex that most people simply leave it alone - but this is becoming more of a problem as more of these devices are introduced in your network, while your users expectations of the network increases.

As we were designing the WLAN unified system, we quickly came to the conclusion that trying to identify the source of "noise" would be nearly impossible for most customers, and this was in fact the impetus for creating our Radio Resource Management (RRM) system. RRM is a set of algorithms that are run to optimize the configuration of the various radio parameters to increase the system's performance (for those that are interested, I will write a piece of RRM in a future blog). RRM certainly solved many of the issues in our customer's networks, by reconfiguring the network to move away from a source that is harming the performance of the network. Ultimately, we need to start thinking about how we can provide more visibility into what is causing the noise - and remove these interferers to help regain some sanity in our air space.

Posted by Pat Calhoun at 06:43 AM Permalink | Comments (8) | TrackBacks (0)

September 10, 2007

What is Enterprise Ready WLAN?

Has anyone noticed the large number of "enterprise class" wireless devices being delivered to the market that only support the 2.4Ghz spectrum?

Whether these devices support 802.11b or 802.11g, the fact remains that there is simply not enough spectrum available in 2.4Ghz space to ensure a quality service. As many of you probably already know, the 2.4Ghz space only supports three non-overlap channels, and no matter how creative your channel plan is, any deployment that exceeds three APs will see some co-channel interference - including from devices that may not be under your control (is your neighbor running a network?).

Backgrounder: co-channel interference means that two neighboring APs share the same channel. When this occurs, if any device in either cell transmits, it is likely to create interference in the other cell. Adjacent channel interference means that neighboring access points are on channels that follow each other. In these cases, physical distance and radio quality can cause some bleed over across the channels and create interference. So in a reason that in an ideal network, you want to ensure that neighboring APs do not share the same channel, and the channel spacing eliminates adjacent interference.

Going back to the 2.4Ghz spectrum, with a total of three channels, you can now see that eliminating co-channel interference is a challenge, while eliminating adjacent interference is simply not possible.

It is for this reason that enterprises need to start considering moving to the 5Ghz spectrum. The number of channels in 5Ghz ensures that sufficient spacing can be established between the APs to eliminate both co-channel and adjacent channel interference. What I would recommend is that in your purchasing decisions you make sure that the devices support 5Ghz. I've heard many arguments from device manufacturers that 5Ghz WiFi is simply too power hungry. I would point those manufacturers to Cisco's 7921 WiFi phone, which has an impressive stand-by and talk-time battery life. We've reached a point where technology is simply no longer a reason for ignoring enterprise spectrum demands.

Many customers have asked whether they should dedicate their 5Ghz spectrum for voice, and push all data to 2.4Ghz. Given the sensitvity of voice service, and the fact that poor performance is much more noticeable to the user, it is a reasonable approach. For customers that deploy dedicated voice devices, such as the 7921 mentioned above, enforcement may be possible. However, for most customers I believe this is an interim step since multi-mode devices are now becoming widely available. These devices provide both Unified Communications and data services, making it very difficult to dedicate a single spectrum for voice - and causing both data and voice on the 5Ghz band.

So if I have both data and voice services on my 5Ghz band, how will I ensure the voice quality meets my user's expectations? As we are seeing through the WiFi certification process, more voice devices are now being certified to be WMM compliant. The WMM certification process provides quality of service (QoS) enhancements to ensure that voice traffic is properly prioritized. I believe that over time WMM will also become an "Enterprise Ready" requirement.

Posted by Pat Calhoun at 10:47 AM Permalink | Comments (0) | TrackBacks (0)

What is Enterprise Ready WLAN?

Has anyone noticed the large number of "enterprise class" wireless devices being delivered to the market that only support the 2.4Ghz spectrum?

Posted by Pat Calhoun at 10:47 AM Permalink | Comments (0) | TrackBacks (0)

September 06, 2007

What's up with MIMO?

With all of the recent talk about 802.11n, I am frequently asked to explain what is MIMO and its benefits. I suspect this may require a few blog sessions, so today I thought I'd focus on multipath and antenna configurations.

So what's up with 3 transmitters and 2 receivers??? Given how much I enjoy music, let me use home audio as an analogy. Remember the days of mono speaker systems (yeah, I know... I'm dating myself somewhat here). Well, in a mono system, the audio system sends all of the audio signals through a single speaker. The sound takes the fastest path to the listener, which I would claim in most home environments would be direct line of sight. While this does provide a full audio experience, there is no directionality in the sound which lessens the entertainment experience. And here is where 5 channel digital audio comes into the picture. In such setups, you will have three front speakers; center, left and right. First off, you will notice a similarity between MIMO and home audio technologies, since there are three front speakers, but most listeners that I know are restricted to two ears. Of course, the question that remains is whether lack of a third ear on our forehead decreases the total experience. These three speakers send independent audio signals, each with directional sensitivity. If a plane is moving from the left to right of the screen, the sound will match.

Unfortunately, that's where the comparison of MIMO and audio systems will end. As with the audio example, an 802.11n access point has multiple antennas, and they are typically referred to in terms of transmitters and receivers. So when you hear about a 3x2 802.11n device, this would mean three transmitters and two receivers. However, unlike the audio system which sends a different audio stream on each front antenna, each transmitter in the 802.11n access point transmits the same signal. So why bother sending the same signal on multiple antennas? Well... here's where WLAN also differs from a typical audio system. Most audio systems are used in a single, typically square or rectangular room, which provides for an optimal experience. However, an 802.11 home network is expected to be available throughout the home, and this requires the signals to traverse through various building materials, including walls and doors. Interestingly enough, much to my mother's dismay, when I was young I also seemed to think audio should be enjoyed in a similar manner.... but that's another story.

Going back to WLANs, office environment have many metal obstructions, including filing cabinets and but also building materials found in walls, ceilings and even door frames. These metalic materials do not absorb the signal and therefore cause it to be reflected, and unlike the human ear, electronics are much more sensitive to reception delays. These reflections are called multipath, which historically would significantly hamper network performance because a single antenna would receive multiple reflected signals at different times - causing receiver interference. However, since an 802.11n device has multiple antennas, the separation between these antennas causes the multiple signals to follow a different path to the receiver, basically using the properties of multipath to its advantage. These signals are then received on multiple antennas, which is where the magic of MIMO comes in. These multiple received signals are recombined into a single stream to increase the strength of the signal.

Well... I gotta run so next time maybe we can talk about what happens when that third ear grows on your forehead.

PatC

Posted by Pat Calhoun at 07:37 AM Permalink | Comments (3) | TrackBacks (0)