Matt Stabeler

Meeting with Paddy – pep talk to make me feel better – Paddy took 3 1/2 years to complete PhD, but changed his subject after tranfer report – and still managed to finish on time – even though I AM NOT brighter than him, I should be ok 😉

OK – lets get going with the city wide environmental sensing!

Need to think about:

What simulators are there? What infrastructure do we need to get it to work?

The underlying hard questions are:

1- Collecting Data

Offloading data in normal operation to fixed nodes or drop off points about the city-

issues

• Storage
• Sampling
• Longevity
• Power consumption
• Date/Time  corroboration

The things on the body communicate, perhaps one is doing the monitoring, and others are managing the storage and dissemination of data across local nodes

Managed the local set as a community

2 – Global Control – Global Phenomena drives local operations

If we can assume nodes have a facility to geo-locate themselved, nodes can use or build a map of the area they are in. As the overall control system (?) detects issues or hotspots, it may increase the frequency of sampling in a particular area – this will require mechanisms to cope with this – considering multiple hot spots.

The idea being that some how nodes would be told to make more detections on certain areas – or even based on certain criteria – e.g if CO2 level > x increase freq.?

This then feeds back to the community – perhaps imagined as a  big async. query over the community.

3 – Cute part

When nodes pass through a hot-spot – what happens if they are not going past any other known drop off  node? Communicate with a node passing the other way through?  How to we manage this and what data – dead-reckoning? probability of passing another node?

Do the nodes always promiscuosly send data – or just sometimes?

Nodes Collaborate to communicate a response and are managed as a community or a swarm

We defined some next actions:

1 – Create  1 page vision document – deadline 3-4 days

To describe –

• Community management
• Fixed Nodes about an space
• Group of devices about a person
• Community of individuals with wearables
• Granularity of resources (?)
• How the overall community will work – what will be the infrastructure  – given our resources

2 – Slick the document up, and send it off to out peers in Harvard, to see if it will fit into their project

3 – When bored of writing – consider:

• How do cities work? where do people go?
• How do we simulate a city wide sensor network like this
• Look at existing applications to simulate cities -flame.ac.uk – metropolis.tcd.ie – Steve Collins and Carol Sullivan
• Look into cheap environmental sensors – bluetooth based – if we can get some which are cheap enough – we can get some and deploy over CASL/CSI/Campus?

4 – Overall Goal is to have a 6 month project – which will support my PhD, to make these new tecniques that we will develop – either as a simulation, real world deployment, or a mixture of both,

• Will need to research the idea of decision theory
• Community resource management

5 – Personal Task

• to find out what research is going on in UCD and Clarity
• Start with Barry Smith – ask who is doing what work – paddy wants to do wearbles – and wants to know who’s doing what research in the area

6 – Look for places to put position papers into

Finally – we need to keep some pressure on – and have some quick turn-arounds – and I should talk go talk to him if the ideas we discussed dont make sense afterwards.

Met with Paddy and discussed what we need to be on with – we both seem to have been thinking about different things – I seem to have gone down a route of something to do with mobile phones – but it needs to be linked to what we are getting at. I said that I don’t really know what wearable computing really means – as I haven’t seen it in the flesh!

Really need to get into the idea of what infrastructure is needed to make wearable systems work – think of it more as network of wearable sensors.

What issues are there when we are thinking about transferring data amongst nodes – what if instead of being centrally managed – what if they could freely communicate with each other.

Discussed the idea of city wide environmental sensing – a mixture of fixed and mobile nodes which are tagging and recording data – offloading to nodes with big pipes – how does this work – and how do we implement it?

Also need to think about issues such as transient and persistant data – what is kept and what is shared and removed – what is private what is encrypted what is not?

Example application is
citywide environmental monitoring – fixed nodes talk each other and share data,
they also talk to mobile nodes to get data –application to monitor an area.

Nodes about the body can
communicate to other nodes on other bodies, handing off data – gossiping –
every piece of data has timestamp – and is tagged so that that it can be
synchronised at a later date.

Also –  talked about  developing some things that we say is good for
wearable computing – i.e. p2p vs managed nodes then compare previous wearable implementations
to our idea of whats best and see if they succeeded.

Hundreds of shimmers
downstairs in clarity to have a play with.

Speak to Julie doyle
about user study – work out what we are trying to get out and keep it in the back of my head.

Paddy suggested this was a turning point – are we
interested in hard computer science part?  Or more the HCI part?

Overall we really get down to the
basics –

What will be good
about P2P vs Central Node?

What is persistant
what is transient?

Matt Walsh – they have
the OS with low level access to data – what do we need to do to get something up and running?

Had a quick chat with Aaron about the survey I am planning, and he suggested picking out a number of busy job types, where people are very mobile (i.e. travelling etc, not just at home with a laptop), and trying to show that somehow using ambient/pervasive technology would improve some aspect of their job – e.g. improved productivity by 5% or whatever – or perhaps a nurse in hospital having an ambient monitor about her person that keeps a track of patients condition.

Also suggested jumping people and asking them about their devices power and usage –

How much power does it have (without looking)? Why do you think that? what applications have been running?

Also talked about opportunistic annexing – using devices to interact with the environment – much like the personal server (Trevor Pering et al)

I’ve been looking at how things like processing power, battery life, and communication bandwidth affect a user on a day to day basis, and have trying to find studies that have measured these things, without success. 

I am planning to do a quick and dirty user study to find out mobile usage patterns: More specifically – 

What sort of devices do users own, what do they use them for, what types of users can we define, what are users familliar with, etc.

Out of this I hope to show that we do not need to optimise for battery power, communications bandwidth is sufficient for most day to day tasks, and that processing power is abundant.

I get very easily distracted with ongoing projects that perhaps I should not be thinking about. To identify and remove any distractions, I listed them out as follows:

• Basadaeir (Website Tweaks)
• Bluetooth Spotter
• Oldtowncacarnivalweek.co.uk(Tweaks, Photos)
• Twitter
• Facebook
• Stabeler.com (Tweaking, Blogging)
• Music
• Olympic Games
• Beer Drinking
• Coffee Breaks
• IM
• Cleaning Desk
• Getting Drinks
• MAC Mini setup
• Online Banking
• Watching Movies/TV
• Worrying about direction of research
• RSS Feeds
• Email!!!!!
• Googe Analytics
• Google Adsense

Most of which are distracting whilst at uni (when I should be concentrating) – so – now I have identified the things that distract me, I should be able to avoid them more easily.

I’ve been reading lots of things recently about wearable computing, but previously I was reading around context and situation awareness, had a number of thoughts based on the ideas I read. Alot of my more recent reading has also referenced some of this material, so it seems  it may be useful, and I have written a short document about how I understand it at the moment.

Ideas About Context

Met with paddy this morning to talk about where to go from here.

I’ve been reading around Context, Privacy, Infrastructure, Management etc. in relation to wearable computing. There are a number of texts that mention it, but nothing (so far) that has dealt with it directly.

Paddy suggested reading and using the idea from http://infoblog.stanford.edu/ that you take the main assumptions about an area, list them out, then change one or two of them, and see what happens. (yet to read artical)

Also, paddy suggested to have a think about stream processing.

I should also look at the locations/conferences/groups that are doing these thing – i.e. find papers that are newer, and may not be highly cited.

Suggested conferences to look into:

MobHoc, MobiCom, Middleware, etc

Paddy also mentioned that reading can expand to fill all available time, so don’t let!

Paddy and I met to discuss the IRCSET Report and my research direction.

Regards the former, the document I wrote was fine, and Paddy will submit a supervisor report separately (as per the IRCSET guidelines).

Regards the latter, I said that I felt that although I had been reading around many areas, and was enthused by attending Pervasive 2008, I did not really have a particular topic that I was readings around. Paddy suggested that allthough I effectively have the freedom (within reason) to research what I wanted, he has some ideas that I could work on.

The topic he suggested was to investigate Wearable Computing/Technology, and to build up my knowledge on the subject by spending the following week doing a breadth-first search of the related literature. We suggested and we agreed that the emphasis could be on the infrastructure and management of wearable technologies within pervasive computing environments. Areas/Issues of interest (some added by me later) being:

power, communications, compliance, privacy, commonalities, categorisation, storage & reporting, processing capacity……

The 1 week project resulting in me finding 3-4 good papers in the field that Paddy and I could use to discuss further research.

Also discussed at the meeting, with Aaron was what to do with the Basadaeir system, we decided that we should meet together (Paddy, Aaron, Matt and Julie) to talk about  where this project could head.

We also briefly talked about Aaron’s Bluetooth idea of encoding communications into Bluetooth name strings, and decided to discuss that at a later date. Aaron will send me his thoughts to date on the project.

In todays supervisor meeting, we discussed mostly a new project, for submission to Ubicomp 2008 (deadline 4th April) which combines sensormash and CFP into a robust system for matching people to interests and bringing about new possibilities for collaboration by displaying information in an untuitive and informative way.

details are here: Meeting 27Feb2008

The last couple of days I’ve been trying to get Ubitracker to run directly from construct. I have created a Contruct class in PHP which deals with the construct communications, for both querying (SPARQL) and inserting data (RDF Triplets).

This enabled me to create an application that uses this to get the data from (any local) construct directly, and format it as XML as expected by the ubitracker interface.

There is more work to be done to do with meta-data needed to run the interface (such as tag/object bindings), once this has been completed, the interface can be run and installed very easily. This would allow other groups who utilize Ubisense, to run this application. A blue sky view of this project would be to create a platform within which sensor ‘plugins’ could be developed and used to display interesting things to users.

Further stages of development will require some rethinking of the output XML structure, to include other sensor types and data (such as Bluetooth readings/locations, Calendar linkage, Webcam images, etc. etc.)

It will be particularly interesting to develop an application which uses all of the location type data to get a good approximation of an objects (persons) location.