Project Assignment - "WhereAreYou?"
Johan E. Bengtsson <email@example.com> Centre for Distance-Spanning Healthcare at Luleå University of Technology
Mobile phones are quickly becoming more capable, with GPS and WiFi support being the norm for high-end phones. RFID readers are soon coming to everyday mobile phones, under the label Near Field Communication (NFC). The MemoryLane project (www.memorylane.nu) aims to support people suffering for mild dementia with information about their life, using People, Places, Objects, Time and Activities as the core concepts. This has the potential to create extremely simple interfaces for personal information management, also for healthy persons. A key aspect of MemoryLane is to allow a relative (or carer) to support the person in their daily lives, by having agreed access tothe liefelog data of the person. During the day a dynamic interface is needed for the relative (carer) to find about where the person is, possibly also in the form of messages when the person enters or leaves pre-defined geographic areas.
Create a proof-of-concept demonstration showing how a mobile phone with GPS and WiFi can detect the current location and make it available to a relative or carer as a web page and messages.
This list of requirements expresses achievements that you should aim for, in order of priority. 1. [DETECT] The mobile should regularly obtain the GPS position and transfer it to a server where it is logged in a database or file. There should be a button to turn off this logging, but this is not mandatory. 2. [MONITOR] After logging in, the carer should be able to view the current location on a map in a web page. 3. [KNOWN PLACES] The carer should be able to define geographic areas as "known places". A simple model using a center point and a radius is acceptable. 4. [ALERT] The carer should be able to set an alert for then user leaves (or enters) a known place (for example home). When this occurs, an SMS is sent to the carer. 5. [WIFI] The mobile should be able to detect the SSIDs, MACs and signal strengths of nearby WiFi bases stations and use this as a "fingerprint" to detect indoor places. Something simplistic is expected here (for example knowing just the WiFi network in the home). Consider autocalibrating by correlating measured GPS positions with detected WiFi networks. 6. [KNOWN PATHS] Provide a learning mode where the mobile learns paths (for example preferred paths leading home), stores it in a suitable format and allows the carer to visualise it on a web page. You should be able to support at least requirements 1-4 during this course.
You will need to implement a client executing in the background on a mobile device with Windows Mobile 6 as the operating system. Accessing lower-level interfaces for GPS and WiFi might not be possible directly from Java, so a C# part might be needed. A server part will be needed, with a store for positions, known places and paths. A simple backend is needed that can generate web pages (after logging in) for visualising the current position, known places and paths. Visualising positions, paths and places in a web page can be achieved through integration with Google Maps. Sending SMS from the server can be done using Clickatel, which provides a number of net APIs for sending SMS to any phone number, using a pre-paid account. Nokia has a project aiming to use WiFi detection for indoor positioning. Iphone apparently already has a similar functionality. Ekahau provides WiFi-based positioning after extensive calibration. You will have access to two mobile phones with Windows Mobile (HTC Touch Cruise with Win Mobile 6 and HTC P3300 with Windows Mobile 5, possibly to upgrade to Win Mobile 6). You will have a Tikitag reader and sticker RFID tags for easily activating the carer webpage on a PC.
   MemoryLane project, http://www.memorylane.nu Near Field Communication Forum, http://nfc-forum.org/ Tikitag PC reader and sticker RFID tags, http://www.tikitag.com