This is an old revision of the document!


Distributed Objects for Real

Real Objects

The widespread adoption of RFID tags provides a unique opportunity to equip physical objects with a digital presence. While the idea was previously explored in the Cooltown project, the introduction of RFID tags as a cost-effective means to tag virtually any object implies a massive increase in scale. It is important to note that some technical problems remain to be addressed in order to allow the user to roam a world populated by a vast amount of objects, both digital and physical.

One problem of note involves the power requirements of an RFID reader. To communicate with passive RFID tags, the reader should emit a magnetic field which provides the RFID tag with sufficient energy to be able to respond. The strength of the field emitted by the reader determines the distance at which the tag can be read. While a fixed reader can continuously scan its environment within a broad range, mobile readers currently operate at a significantly smaller range in an effort to conserve battery power. For instance, our fixed Feig HF reader has a maximal reading range of 18cm (with other HF readers boasting ranges of 30 to 60 cm) as opposed to a mobile Cathexis Bluetooth reader which only detects tags within 1 cm. However, it is conceivable that in the near future mobile readers will be able to scan tags in a considerably broader range.

Anticipating the advent of mobile readers with an extended range of operation, the authors believe that it is crucial to devise mechanisms that allow application developers to deal with a situation where users carry mobile devices capable of detecting physical objects in their environment. A thorough discussion is needed on how to model a collection of nearby objects which is constantly in flux. Over the past year, we have developed two opposing programming models to represent RFID tags in an application. In this position paper we present these models as two extremes in hopes of inspiring discussion on how mobile applications should deal with a world populated by both digital and physical objects.

Prototype implementations of both models have been developed in AmbientTalk, an actor-based domain-specific language built on top of J2ME. In this position paper, the models are explained as independently as possible of their realization in AmbientTalk.

Volatile Data Clouds

The volatile data clouds model considers an RFID tag as data whose presence or absence can be used to steer the application. Therefore, it provides an explicit representation for the collection of visible tags. This collection is implicitly tied to the environment and is likely to change frequently. Such changes are at the very heart of the model and consequently application programmers do not need to manually specify observers to react to them. Instead changes are propagated implicitly and automatically.

The implicit and automatic propagation of change is achieved by borrowing ideas from functional reactive programming, which introduces reactive values. Changes to these reactive values (which represent e.g. time, the speed of a robot wheel etc.) are propagated to a network of dependent computations. Such a network of dependent computations is built when functions are applied to one or more reactive values. The result of applying such a function is a new reactive value, which can in turn be used to construct additional dependent computations.

When programing with reactive values that represent the collection of visible tags, the programming language needs to provide the following support. First of all, support is needed for reactive collections, such that operations such as size and contains result in a reactive value that changes when the collection changes. This necessitates the introduction of reactive programming support in the programming language. Reactive programming has been successfully incorporated in various languages including Scheme, Java.

Furthermore, such collections should provide memoized iteration constructs (e.g. map:). Hence, when mapping a function f over a source collection s, a memoized map: produces a derived collection d by simply mapping f over the current elements of s. If at any later point in time, a new element a is added to s, the update will not cause the derived collection to be computed anew. Instead, d is extended with a single new element, to wit f(a). Similarly, when an element r is removed, only the value is removed from d.

Most importantly, the programming language should provide support to detect interesting patterns as they occur in a reactive collection. A pattern of interest can be the presence of a particular (kind of) tag, but should extend to combinations of interesting tags. Therefore, we propose a model which uses pattern matching rules to detect changes of interest in the application’s environment. An application could for instance specify a rule which triggers when the tags that represent the user’s keys and coat are detected simultaneously. Such a rule could be used to remember that the user is likely to have left his keys in one of the pockets of his coat.

Tags Objects

While the volatile data clouds model considers RFID tags as containers of data which is to be filtered and interpreted by the application, the tag objects model interprets them to be full-blown objects. Applications can interact with these objects and e.g. invoke methods on them. In this case, the tag contains a marshalled object representation which is implicitly unmarshalled to provide a live object which applications can interact with.

When treating tags as objects, it is important to deal with the ephemeral nature of the connection between the mobile application and any particular tag. A first problem to be tackled is how applications detect new tags as they come into range. This is addressed by the introduction of a semantic service discovery mechanism, which allows applications to be notified when an interesting tag appears. The use of a service discovery mechanism also provides an implicit mechanism to disregard tags which contain objects that are irrelevant to the application. For instance, an application can choose to only discover objects of type Book or Ingredient.

Once a tag object has been discovered, the application can start to interact with it. However, if either the user of the application or the tagged object is roaming, it is extremely likely that the tag will (temporarily) go out of range. The tagged object model requires that the programming abstraction are sufficiently robust to deal with such sudden disappearances. Therefore, tag objects should be treated as “remote objects”; objects which can only be addressed using asynchronous messages. Hence, when a tag goes out of range, all messages that are sent to it are buffered. Subsequently, when the tag comes in range again, all buffered messages sent to it can be flushed.

While messages are implicitly buffered during a temporary disconnection, it is important to provide the application programmer with the necessary abstractions to explicitly detect when a tag has disappeared or when it has been unreachable for a certain period of time. Therefore, the programming model should allow the application to explicitly provide disconnection and reconnection listeners.

Within the tag object model, we have conducted initial experiments on how to use ad hoc replication strategies which could allow continued use of a tag object even when the tag is temporarily unreachable. When the tag becomes reachable again, updates can be synchronized. With such a custom replication strategy in place, it can in principle be possible to treat tag objects as “local objects”; objects which can be addressed with both asynchronous and synchronous messages.

Comparison

Having implemented both models, it remains unclear whether one of the models presented in this paper is to be preferred over the other. As it stands, both models have shown to cater to different kinds of applications. Moreover, when analysing the strengths and weaknesses of each of the models, they have shown to be largely complementary.

The volatile data clouds model treats RFID tags as simple containers of data and aims at providing applications with expressive means to represent a collection of nearby tags which is constantly in flux. Consequently, the model focusses on being able to expressively filter such collections and detect the presence of (a combination of) tags, to which the application can respond. Using volatile data clouds as a means to represent nearby tags allows the programmer to abstract over the events that underly the discovery of new tags and the handling of their disappearance. The cost incurred by using these abstractions is the need for a rather heavy-weight infrastructure to properly handle reactive computation. Furthermore, this model considers RFID tags almost exclusively as a means to indicate the presence of a physical object in close proximity to the device that is running a particular application.

The tag objects model on the other hand treats the contents of an RFID tag as a full-fledged object. This makes the model capable of interacting naturally with tags that contain marshalled objects and makes it particularly interesting to model interactions with active tags. Active tags are RFID tags which contain their own power source and are capable of independently performing (some limited form of) computation. Another advantage is that the model only requires a service discovery mechanism and a means to buffer messages, which makes it considerably more light-weight that the volatile data clouds model. The downside of this method that it provides only crude support to detect the presence of (combinations of) tags in the environment. Furthermore, the application needs to explicitly provide listeners to react to the events signalling the disconnection or reconnection of a tag.

In all likelihood, a programming model that fully leverages the advantages of a world teeming with tagged objects will incorporate elements of both models presented here.

Further Reading

* Distributed Objects for Real. Stijn Mostinckx, Andoni Lombide Carreton, Kevin Pinte, Wolfgang De Meuter. Technical report, 2010, Vrije Universiteit Brussel pdf

research/doforreal.1280860177.txt.gz · Last modified: 2010/08/03 20:48 by stijnm
 
 
 
Recent changes RSS feed Creative Commons License Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki