Sensor Networking Is A Challenging Research Information Technology Essay
In sensor network, the system is required to produce a desirable outcome under a dynamic environment with high degree of uncertainty, noise, faults and other uncontrolled factors. With more demanding requirement from a real-time system, new paradigm shift is needed. Wireless sensor network normally has few resource constraints such as limited battery power, computing power, and limited bandwith links among sensor nodes. Thus, sensor network is not suitable for sophisticate protocol and the inter-communication among sensor nodes is also limited. Compared with ad-hoc network, there are large numbers of densely distributed nodes. Therefore, we need to find the best way to compute density for each situation to keep the density and power at minimum to cut down the costs for the system. Other important challenges mentioned are real-time and security.
The network challenge lies in MAC layer design. Some MAC protocols are mentioned such as Scheduling based MAC protocols, Collision Free Real-time MAC. Scheduling based MAC protocols, TDMA has limited scalability and adaptability because it is hard to dynamically change the frame size or slot assignment when new node joins in. Some TDMA protocols still have collision. It is difficult to use TDMA in sensor networks.
For collision free real-time, the schedules are kept identical so that each node knows exactly which other node has the message with the shortest deadline and has the right to transmit. It can be done by replicating EDF schedule at each node for packet transmission. It is shown that implicit EDF provides higher throughput even during workload. It saves power by avoiding collisions, but poses multi-channel problem.
In contention-based MAC, node changes are easily accommodated and multi-hops communication are supported. Examples of contention-based MAC are DCF, MACAW(improved over MACA). DCF uses binary exponential backoff, allows physical and virtual sensing, broadcasts packets are sent directly after carrier sensing. Because of back-off nature, MAC does not guarantee the priority order of packets from different nodes.
Existing MAC protocols do not adequately consider the requirements of sensor networks.
Many excellent protocols have already been developed for ad-hoc networks. They can be categorized into two groups: (1) flat routing and (2) hierarchical routing. In flat routing, all routes have equal responsibility for maintaining the routing information. Routing algorithms in this category can be further classified into three groups: (1) Proactive, (2) Reactive, and (3) Geographical. Proactive routing algorithms maintain routes continuously for all reachable nodes. They usually require periodic dissemination of routing updates. Reactive routing algorithms establish and maintain nodes only if they are needed for communication. New routes are acquired when a new connection is set up and is to be maintained throughout the lifetime of connection regardless of topology changes. Geographical routing protocols utilize location for routing decisions. Hierarchical or cluster routing is the procedure of arranging nodes in a hierarchical manner providing end-to-end real-time guarantees is a challenging problem in sensor networks. SPEED is an adaptive, location-based real-time protocol aim to reduce end-to-end deadline miss ratio.
The amount of needed state information and the signalling overhead, reservation schemes are unlikely to scale well in sensor network
Transport layer often result in deteriorating throughput and fairness in ad hot network due to fairness of the underlying MAC protocol and link failure due to mobility , coupling effects of the forward and reverse paths. But there exists little work on the transport layer issues in sensor networks. In sensor networks, usually there are few sinks to which packets are directed and experimental results and analysis have shown that in-network aggregation can achieve lower energy consumption and higher data delivery raito than address-centric communication
Topology control deals with reduce topology of the network to save energy. There are some solution proposed in COMPOW that is a topology control protocol or LMST is a topology control algorithm to preserve the network connectivity
Power management: there have been several energy conserving protocols for ad-hoc networking environments. Take advantage of route redundancy and turn off radios that will not affect network connectivity. A common approach is to construct an overlay backbone composed of a small number of active nodes to route all multi-hop packets, while let the other nodes sleep when they do not send or receive data. LEACH adopted a hierarchical strategy. GAF also allowed a hierarchical strategy, but its clusters are based on geography.SPAN forms overlay backbone in a peer-to-peer fashion.
Real-time is another issue that crosses all layers in the communication stack. RAP is a multi layer real-time communication architecture for sensor networks. Communication on RAP is addressed by location. Communication in RAP is supported by a scalable and efficient protocol stack, which integrates a transport layer location addressed protocol, a geographic routing protocol, velocity monotonic scheduling layer and a contention-based MAC that supports prioritization.
Operating system: distributed computing brought about by the advent of sensor network requires revisiting the basic operating system abstractions such as tasks and inter-task communication as well as developing support for fundamentally new distributed programming environments. Sensor networks need the integration of objects that live in physical time and space as components in the computational environment of the application.
Lowest system support for sensor networks begins at the level of the single node. tinyOS is perhaps one of the earliest operating system kernels developed exclusively for sensor nodes.
Distributed virtual machines to provide convenient high-level abstractions to application programmers while implementing low-level distributed protocols transparently in an efficient manner.
Context awareness: Sensors are aware of their context, neighbouring and history.Given the context, sensor can predict their reading. It is a challenge for sensor networks to develop a full vision of context awareness
Content-addressable space is required in sensor network so that destinations are addressed by their contents attributes, not by their machine identities. It is challenging to find the efficient way of data propagation.
Open problems such as distributed control, group management, team formation, data services
In conclusion, this paper discusses the state of art of sensor network, addresses its challenges as well as some open problems.
If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal: