Wireless Mesh Networks: What is it?
I have decided to open this category with a post on something I am personally doing research on. I will call this category ProjectWM. These are just random notes. But, hopefully, I can get some people interested enough to contribute a discussion or two in this blog. But, if you are shy then I can settle for email. So on with the opening salvo. Welcome!
Whenever we hear the words “Wireless Mesh Networks (WMN)” different images come into mind. Some of think of it in terms of telemetry networks of distributed sensors and others think of it as a last mile networking technology for providing Internet access. What does this term aim to describe? Well, the truth is the term WMN can be used for a number of things including the two (2) mentioned above. But, before that. Let us try to come up with some key criteria for calling something a WMN [3][4].
- Wireless. Yes. The most important property of a WMN is that it is wireless. It does not require any physical cabling to provide network connectivity. This allows network builder to save on time and resources.
- Ad-Hoc. This allows network to be created on the fly. Since, the nodes are connected wirelessly it is possible to do this cost effectively.
- Distributed. All this connectivity must be provided for a purpose right? Being able to have nodes positioned in various locations across a given geographical area. There are many uses of a WMN and being distributed in nature contributes to these functional requirements.
- Mobile. This is one of the potential benefits of being wireless. The nodes can potentially be moved around. However, there are two (2) generally types of mobility. Let us call one nomadic. This allows nodes to change locations and maintain connectivity after changing locations. The other one we can call fully mobile. This allows nodes to maintain connectivity in the process of changing locations. Nomadic users only use the network when it has been setup. Examples of nomadic networks are the Motorola Canopy system (pre-WiMAX) and WiFi systems (which may not necessarily be mesh networks). While, full mobility users want to be able to use the network on the go. Examples of full mobility networks are our cellular networks which consider issues like handover (but are not mesh networks).
- Connected. WMNs being networks must connect things together. This connectivity is used to exchange or relay data amongst each other. This is the reason for being of any network.
Then, let us move to some of the more important uses of a WMN. The central use of this technology is to provide network connectivity. This network connectivity can have different uses. These uses can be divided into the two (2) main groupings below.
- Provide network access to collect information via distributed telemetry. Remote sensors can be placed in various locations and connected using a WMN [4]. This is a cost effective way to network various sensors without having to invest (time and resources) in cabling. This also allows the network to created and removed quickly for highly portable projects. It would make sense to re-use sensors in other areas after their initial use. Being wireless provides this flexibility.
- Provide network access to distributed mobile users. This is primary reason why MIT initiated the Roofnet project [1] and other similar projects [2]. This connectivity can be used to provide distributed Internet access to a community or neighborhood using WiFi technology. Or it can be used to connect battlefield or public safety computers quickly [3].
So what is it in for us? Well portability, mobility, cost-effectivity and many others. But, this is the topic of discussion for another blog post. This seems to be a very rich field for research work. I can immediately see a number of open problems that can definitely use more elaboration and discussion. Some open problems have to do with routing, survivability, scalability, security, provisioning, hand-off and many more [3][5]. Again, each merits a discussion on its own.
References:
[1] D Aguayo, J Bicket, S Biswas, D De Couto. 2003. MIT Roofnet: Construction of a Community Wireless Network.
[2] R Bruno, M Conti, E Gregori. 2005. Mesh networks: commodity multihop ad hoc networks. IEEE Communications Magazine.
[3] I. Akyildiz, X. Wang. 2005. A survey on wireless mesh networks. IEEE Communications Magazine.
[4] S. Rupley. 2003. Wireless: Mesh Networks. PC Magazine.
[5] M. Sichitiu. 2005. Wireless Mesh Network: Opportunities and Challenges. Proceedings of World Wireless Congress.
April 3rd, 2007 at 8:56 am
[…] After my previous blog entry about WMN and after a bit more reading, two (2) additional important properties emerged that add to the definition of wireless mesh networks (WMN). There two (2) properties are described in brief below. […]
April 17th, 2007 at 1:45 am
[…] These types of networks will have many of the characteristics of our previously featured ad-hoc wireless mesh networks such as being wireless, ad-hoc, distributed and connected. The most common uses of wireless sensor networks are commonly used for a whole multitude of applications such as telemetry for both military and civilian applications. The ability to easily distribute, collect data and manage sensors open up numerous possibilities for grown in this space [1]. However, they are not necessarily mobile, multi-hop and multi-channel. Most of these sensors have fixed positions after they are deployed and discovered. Data is typically not routed between the nodes themselves but directly to a central collector. Since, majority of the applications are network telemetry, this does not require a lot of bandwidth and thus does not require a whole lot of channels. […]