[thought] On Bluetooth research.

Problem Area 1: Bluetooth

 

When bluetooth enabled devices are brought close to each other, they automatically communicate to form piconet?s. A piconet consists of one master and upto 7 slaves and any bluetooth device can function both as a master as well as a slave. The master is the central entity that decides transmit/receive-slot allocation to different slaves and thus controls bandwidth usage among slaves. Thus it is important to chalk out such role for different bluetooth devices, as transient devices (like a PDA or a laptop) are more suitable to be slaves than masters (like a desktop computer). Every piconet is characterized by a unique frequency hopping sequence code that is used to communicate between the master and the slave. NOTE THAT: the distance between the bluetooth devices does not influence the piconet formation. It is entirely possible for bluetooth devices, which are placed close to each other to be in different piconets. The following are some important research issues:

 

 

·        Initial Synchronization: To form piconets, the devices need to agree on some particular hopping code sequence and need to synchronize with each other. Since there is no common single frequency that a device can listen to (or transmit on) to communicate with other devices, an open research issue is to find techniques to allow quick synchronization between two or more devices and assigning roles for masters or slaves to each such device.

 

·        Optimum Topology Construction: Two or more piconets can be linked together to form a scatternet. In this, one device in each piconet (either a Master or a slave) acts like a bridge between them, relaying information from devices in one piconet (using one frequency hopping code sequence) to another piconet (using another frequency hopping code sequence). An open research issue is how to form such scatternets and what will be the most optimum topology to form such scatternets. For example, the following problem might motivate some thinking into this area.

o       There is a desktop PC, which is equipped with bluetooth. Assume that it has already formed a piconet with a mouse and is communicating with it. A table PC with a pen, elsewhere has already formed another piconet. When the tablet PC is brought near the desktop PC, what will be the most optimum topology? Either having one piconet with the desktop PC as the Master, or having a scatternet with each PC acting a master in his own piconet, and any of them acting as a bridge between the two.

 

·        Quality of Service issues: Bluetooth devices can support a maximum bandwidth of 1Mbps (under most ideal conditions). If bluetooth is to be expected to be ubiquitous then it better support both quality of service as well as higher Bandwidth as multimedia applications require both. For example, consider the case of digital cameras that can capture motion video (possibly in MPEG format). If it is to be connected to a PC using bluetooth, then it is necessary that it be given certain guarantees about the bandwidth without affecting other devices. In this case, a desktop PC and a digital camera might form just a single piconet, and disallow any other device to particpate in that piconet due to lack of bandwidth. (Some form of admission control might become necessary).  

 

·        Power Efficientcy: It is anticipated that most bluetooth devices will be computationally impoverished and have limited source of power. Hence any solution devised for the above research problems will require high power efficiency.