Wsn state-centric programming
- 1. State-Centric Programming Done by Aanchal Kumari 17001UEL01
- 2. CONTENTS 2Wireless Sensor Networks Introduction Collaboration Groups Examples of Groups Geographically constrained group N-hop neighborhood group Publish/Subscribe group Acquaintance group Mixing
- 3. INTRODUCTION • CSIP applications, such as target tracking, are not generic distributed programs. • Deeply rooted in these applications are the notion of states of physical phenomena and models of their evolution over space and time. • A distinct property of physical states, such as the location, shape, and motion of objects, is continuity in space and time. • The sensing and control of these states are typically done through sequential state updates. 3Wireless Sensor Networks
- 4. xk+1 = f(xk, uk) (1) yk = g(xk, uk) (2) x : state of a system u : inputs y : outputs k : integer update index f : state update function g : output observation function • This formalization is broad enough to capture a wide variety of algorithms in sensor fusion, signal processing, and control. 4Wireless Sensor Networks
- 5. • Certain concerns were addressed where and when, rather than how, to perform sensing, computation, and actuation, and play a central role in the overall system performance. • However, traditional programming models and languages don’t support these “nonfunctional” aspects of computation well. • State-centre programming aims at providing design methodologies and frameworks that provide meaningful abstractions for these issues. • One such abstraction in this section is collaboration groups. 5Wireless Sensor Networks
- 6. COLLABORATION GROUPS • A set of entities that contribute to a state update. • These entities can be physical or logical sensor nodes or more abstract system components such as software agents. • A group encapsulates two properties: its scope to encapsulate network topologies and its structure to encapsulate network protocol. • The scope of the group defines the membership of the nodes with respect to the group. • A group’s structure defines the roles each member plays in the group and thus the flow of data. 6Wireless Sensor Networks
- 7. • Formally, a group is a 4-tuple G = (A, L, p, R) Where, A is a set of agents; L is a set of labels, called roles; p : A→L is a function that assigns each agent a role label; R ⊆ L × L are the connectivity relations between roles. • Given the relations between roles, a group can induce a lower level connectivity relation Eamong the agents such that ∀a, b ∈ A, if (p(a), p(b)) ∈ R, then (a, b) ∈ E. 7Wireless Sensor Networks
- 8. EXAMPLES OF GROUPS Protocol example: Geographically constrained group N-hop neighborhood group Publish/Subscribe group Acquaintance group Mixing 8Wireless Sensor Networks
- 9. Geographically constrained group: • A GCG consists of a set of nodes located within some fixed geographical region (the group’s extent). • Physical signals, especially from point targets, may propagate only to a limited extent with distance, this type of group naturally captures the set of sensors that can possibly sense a local phenomenon. • GCG’s geographic shape can be specified as circles, polygons, and their unions and intersections. • GCG can be established by geographically constrained flooding, using protocols such as Geocasting7 and GEAR to support communication among members. • A GCG may designate a leader, which fuses information from all other group members. 9Wireless Sensor Networks
- 10. N-hop neighborhood group: • An n-HNG has an anchor node and defines the all nodes within n communication hops are members of the group. • Local broadcasting can be used to determine the scope. • The anchor node is the group leader, having a tree structure with the leader as the root to optimize communication. • If the leader’s behavior is decomposed into sub-operations running on each node, then the tree structure also provides a platform for distributing the computation. • There are several useful special cases for n-HNG. • For example, 0-HNG - the anchor node itself, 1-HNG - anchor node’s one-hop neighbors, and ∞-HNG contains all the nodes reachable from the root. 10Wireless Sensor Networks
- 11. Publish/Subscribe group: • A group may also be defined more dynamically, such as all entities that can provide certain data or services or that satisfy certain predicates over its observations or internal states. • A PSG consists of one or more consumers expressing interest in specific types of data or services and the producers that provide those data or services. • Communication among members of PSG are established via rendezvous points, directory servers, or network protocols such as directed diffusion. 11Wireless Sensor Networks
- 12. Acquaintance group: • A member belongs to the group because another group member “invited” it. • A member may quit the group without other member’s permission • An AG might have a leader serving as the rendezvous point. • When the leader is also fixed on a node or in a region, GPSR (greedy perimeter stateless routing), ad hoc routing trees, or directed-diffusion types of protocols can facilitate communication between the leader and the other members. • This group is to monitor and control mobile agents from a base station. • When all group members are mobile and no central leader exists, maintaining group connectivity might become quite complex. 12Wireless Sensor Networks
- 13. Mixing: • Mixing and matching group is a powerful technique to tackle system complexity, by making algorithm much more scalable and resource efficient without sacrificing. • Using highly tuned communication protocols for a specific application to reduce latency and energy cost. • There are various ways to compose groups. • Parallel - to provide different types of input for a single computational entity. • Hierarchical – it is composed such that one group contains another (or its representative member). 13Wireless Sensor Networks
- 14. Thank You!! 14Wireless Sensor Networks