Pervasive Pheromone-Based Interaction with RFID Tags” by Marco Mamei and Franco Zambonelli is a more theoretical work about coordinating activities with objects by leaving little informations on distributed memory aka digital pheromones realized by RFID tags. They distinguish between object tags and location tags. The actual algorithm is a little puzzled out. In short they increment so called hop counters and timestamps, which are stored in the location tags and compared with the ones in the object tags passing by. A critical issue for their system to work, is a fitting evaporation threshold of the stored information. Further it is fundamental how the ratio between the number of tracked objects and the tag storage capacity works out. They studied their algorithm within a computer simulation and some remote toy cars on RFID tag grid. The tested algorithms were proactive vs. parasitic diffusion. A few interessting statements are:

…the number of deployed systems exploiting digital pheromones to coordinate the activities of situated autonomous agents is still very limited. In this article, we present a simple low-cost and general-purpose implementation of a pheromone-based interaction mechanism for pervasive environments. This is realized by making  use of RFID tags to store digital pheromones and by having humans or robots spread/sense pheromones by properly writing/reading RFID tagspopulating the surrounding physical environment.


The usefulness of evaporation depends strongly on whether agents mostly search recently moved objects (in which case a low evaporation threshold is better) or objects which have not moved for a long time (in which case, a high threshold would be better so as not to delete old yet relevant pheromones).

In another paper by H. Lui et al., they talk about the oppertunity in using multihop networks to extend the range of passive RFID networks. In combination with the RFID powder I already wrote about and the following scenario this would have mad ubiquitous potential:

One could generally think of exploiting RFID pheromones to enable a group of users and robots to coordinate their movements in an  unknown and challenged environment on-the-fly without any  advanced planning. As a practical example, consider an emergency  rescue team (whether human, robotic, or mixed) arriving in a disaster area where no computing/network infrastructure is available other  than the (nearly unbreakable) RFID tags around. On the one hand, if the team members exploit these tags to spread pheromones around as  they walk and are instructed to stay away from existing  pheromone trails, then one can have reasonable guarantees that the  whole environment is explored in a comprehensive and effective way by the group [Svennebring and Koenig 2004]. On the other hand,  whenever a member of the rescue team discovers something important that needs to be conveyed to other members of the team  (e.g., a robot finds an injured person requiring medical assistance), it  can start spreading a pheromone leading to that something so that  other members (e.g., first-aid doctors passing by) can notice it.


Furthermore, the future developmentof plastic (and printable) RFID technology [Collins 2004] allow envision the possibility of enriching an RFID reader with an RFID printer to dynamically print RFID tags on pavements, walls, or any type of surface, whenever needed.


Consider the case of a visually-impaired person having a short-range RFID reader mounted on his white stick, in order to sense pheromone  trails stored on RFID tags attached to the pavement. In this case, the  ability of accessing information stored directly in the environment is  dramatically important and useful and, while such information can be fruitfully complemented by additional information stored in some databases and accessed via WiFi, it can hardly be fully replaced by it.

Further they refer to an relevant article about an 4 megabit RFID chip by HP.