FIELD OF THE DISCLOSURE

The present disclosure relates to the field of wireless sensor networks.

BACKGROUND OF THE INVENTION

Recent advances in the fields of Very Large Scale Integration (VLSI) and Micro Electro Mechanical Systems (MEMS) led the research and development of wireless sensors to monitor natural and artificial phenomenon. Typically, a wireless sensor includes sensing units, a processor, a memory, a transceiver and a battery that gives it the capability to perform operations like sensing, computation and communication with other nodes and a base station to form a wireless sensor network.

The wireless sensor networks are often used in applications such as monitoring, for instance monitoring of forest fires, where wireless sensors are typically thrown from a helicopter into forests. The wireless sensors self-organize themselves and communicate the measurements to the base station/sink. The communication of measurements to the base station is carried out using routing techniques like flooding where all sensor nodes perform the task of sensing data variables and propagating the sensed variables based data packets in all directions for forwarding to the base station. This approach is not energy efficient as it requires every sensor node to sense and transmit data packets and also leads to transmission of redundant data packets to the base station. Thus, it is necessary to develop an energy efficient routing technique to communicate the measurements from the sensor nodes to the base station.

Further, as the wireless sensors are randomly distributed in the area that is to be monitored, it becomes essential to determine the location of a wireless sensor in case any event is detected. For instance, in case the sensor nodes sense a fire over a particular region in the area it is necessary that the exact location of the fire is known to immediately address the detected event. Thus, the problem of localization is very crucial and there are some known approaches like embedding Global Positioning Systems (GPS) in the sensor nodes. This approach, however, increases the cost per node and hence is not cost effective and in addition requires additional node battery power for operation of the GPS device. Hence, there is felt a need for an efficient and energy aware localization technique.

Apart from routing and localization these days spectrum scarcity has also become an issue in the context of communication done by wireless sensor networks. Contemporary wireless sensor networks communicate over unlicensed Industrial Scientific and Medical (ISM) band (2.4 GHz-2.5 GHz). The rapid proliferation of low cost wireless applications in unlicensed spectrum bands has resulted in spectrum scarcity in those bands. Thus, the sensor nodes have to continuously keep sensing for availability of a white space in the spectrum to initiate communication with the base station. Searching for a free channel continuously with respect to time, band and space utilizes a significant amount of energy and reduces the lifetime of the network. Typically, the problem of all nodes sensing the spectrum continuously for transmitting data packets can be overcome by employing a divide and rule based approach. In this approach distributed sensing is performed wherein chunks of the whole band are distributed among several nodes. However, this approach also utilizes a significant amount of energy and requires the designated nodes to scan the allotted bands for sensing white spaces and the share the information amongst other nodes.

Furthermore, most applications in Wireless Sensor Networks (WSNs) utilize the unlicensed spectrum, network-wide performance of WSNs inevitably degrade as their popularity increases. Statistics show that as against the unlicensed spectrum, the licensed spectrum is comparatively under-utilized and can be used as a promising solution to overcome the spectrum scarcity issue. However, the unlicensed users or Secondary Users (SUs) have to opportunistically use the licensed bands without interfering in the operation of licensed users or Primary Users (PUs). Hence, there is felt a need for channel switching techniques which are energy efficient and at the same time enable the nodes to switch to other available channels in the spectrum on detecting presence of a PU in the currently used band.

Thus to overcome the aforementioned drawbacks there is felt a need for a wireless network system which:

• provides a cost effective approach for localization of sensor nodes in a WSN;

• provides energy aware routing techniques which overcome the disadvantages of pure flooding based routing techniques;

• reduces the spectrum sensing time for sensing availability of a white space/ spectrum hole;

• provides efficient channel switching techniques to opportunistically use the licensed spectrum bands; and

• significantly reduces the energy consumption of the sensor nodes to increase the lifetime of a WSN.

OBJECT OF THE INVENTION

It is an object of the present disclosure to provide a cost effective approach for localization of sensor nodes in a wireless sensor network.

It is another object of the present disclosure to provide energy aware routing techniques which overcome the disadvantages of pure flooding based routing.

It is still another object of the present disclosure to reduce the spectrum sensing time for sensing availability of a white space/ spectrum hole for communication.

It is yet another object of the present disclosure to provide efficient channel switching techniques to opportunistically use the licensed spectrum bands.

One more object of the present disclosure is to significantly reduce the energy consumption of the sensor nodes to increase the lifetime of a wireless sensor network.

SUMMARY OF THE INVENTION

In accordance with this disclosure there is provided a system for monitoring natural and artificial phenomenon in an area, the system comprising:

a plurality of reconfigurable, wireless sensors nodes randomly distributed in the area and arranged in a set of clusters, wherein the sensor nodes are adapted to elect and send sensed variables to a cluster head, the cluster head are adapted to continuously sense spectrum availability for a preconfigured number of channels and are further adapted to aggregate and transmit sensed variables based data packets through an available channel using data-centric routing; and • a base station adapted to control and communicate with the wireless sensor nodes and receive the data packets, wherein the base station further adapted to facilitate creation of the set of clusters by partitioning the area into a plurality of levels and sectors using incremental power broadcasts and localizing each of the wireless sensor nodes in the set of clusters based on the level and the sector information sent in the power broadcasts.

Typically, the sensed variables are selected from the group consisting of interval valued variables and fuzzy variables.

Preferably, the wireless sensor nodes are cognitive radio wireless sensor nodes, wherein the cognitive radio wireless sensor nodes include a cognitive engine adapted to employ supervised learning techniques to perform spectrum sensing and radio frequency environmental parameters' optimization based on spatial, temporal and traffic related parameters.

Further, the wireless sensors nodes include an optimized RF front-end unit adapted to dynamically tune its radio frequency and transmission characteristics based on predetermined inputs.

Still further, the cluster head is further adapted to perform data-centric routing, wherein the cluster head is adapted to accept data packets sent from a wireless sensor node and/or cluster head located in a higher level and in a same sector or in a sector which is one hop apart above or below the cluster head.

Additionally, the cluster head is still further adapted to perform data-centric routing, wherein the cluster head routes data packets received from sectors one hop apart either above or below from the cluster head.

Furthermore, the data packet includes location information (level and sector) of a wireless sensor node, which sent the sensed variables in the data packet.

In addition, the base station is adapted to use hop-count based approach to partition the area into a plurality of levels in the event that the wireless network experiences fading.
In accordance with this disclosure, there is provided a method for monitoring natural and artificial phenomenon in an area, the method comprising the following steps:

• distributing a plurality of reconfigurable, wireless sensors nodes randomly in the area;

• partitioning the area into a plurality of levels and sectors using incremental power broadcasts by a base station and localizing each of the wireless sensor nodes based on the level and the sector information sent in the power broadcasts;

• arranging the wireless sensors nodes in a set of clusters, wherein the sensor nodes are adapted to elect and send sensed variables based data packets to a cluster head;

• accepting and routing the data packets at a cluster head if the data packets are sent from a wireless node located at a higher level and in a same sector or a sector which is one hop apart above or below the cluster head; and

• continuously sensing spectrum availability at a cluster head for a preconfigured number of channels, aggregating and transmitting sensed variables based data packets through an available channel to the base station.

Typically, the step of continuously sensing spectrum availability includes the following steps:

• employing supervised learning techniques for sensing free channels for transmission and reception based on spatial, temporal and traffic related parameters and performing radio frequency environmental parameters' optimization for reconfiguring physical layer parameters for getting maximum throughput; and

• dynamically tuning radio frequency to a countable finite number of channels using an optimized RF front-end unit for performing channel switching.

BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS

The present disclosure will now be described with reference to the accompanying drawings, in which:

FIGURE 1 illustrates an exemplary diagram of a system for monitoring natural and artificial phenomenon in an area using a doubly cognitive wireless sensor network in accordance with this disclosure;

FIGURE 2 illustrates a schematic of the cognitive engine embedded in the wireless sensor node in accordance with this disclosure;

FIGURE 3 is a graph showing the spectrum holes (white spaces) and occupied bands; and

FIGURE 4 is flowchart showing the steps involved in implementation of a doubly cognitive wireless sensor network for monitoring natural and artificial phenomenon in an area in accordance with this disclosure.

DETAILED DESCRIPTION

The drawings and the description thereto are merely illustrative of a system to monitor natural/artificial phenomenon in an area using doubly cognitive wireless sensor networks and only exemplify the disclosure and in no way limit the scope thereof.

Contemporary wireless sensor networks are faced with challenges in conserving sensor node energy and efficiently handling spectrum scarcity concerns. Additionally, these wireless sensor networks employ non-energy efficient techniques that treat the routing, fusion and localization problems as being isolated. Also, the measurements from sensor nodes are currently considered as being crispy rather than considering them as being interval valued and/or fuzzy (variables).

The above shortcomings of contemporary wireless sensor networks led the present invention to envisage a system to monitor natural/artificial phenomenon in an area using doubly cognitive wireless sensor networks. The proposed Doubly Cognitive Wireless Sensor Network (DCWSN) provides energy aware techniques which perform localization, routing and fusion in a combined fashion and using data-centric techniques. The DCWSN proposes clustering of the area of deployment so that data packets are not flooded to all sensor nodes in the area but follow a directional / hierarchical path to reach a base station. This approach not only saves energy as only a set of nodes participate in the transmission but also ensures that the base station does not receive redundant data packets.

Further, the proposed DCWSN also provides an effective and energy efficient local positioning technique which performs localization and sectoring in the same way in which the light houses direct ships through a beacon. Thereby, each sensor node in the network can be localized using the sectoring and leveling performed on the area of deployment.

Still further, the proposed DCWSN reduces the spectrum sensing time by embedding a cognition engine in the sensor nodes. The cognition engine senses spatial, temporal dimensions as well traffic patterns in an area where the network is deployed to suggest bands for conducting spectrum sensing. The cognitive approach guarantees good quality communication among WSN and also leads to reduction of battery power consumption. The cognitive engine also performs radio frequency environmental parameter optimizations to enables the wireless sensor node transceiver to dynamically reconfigure its parameters to gain maximum throughput and arrive at good Quality of Service (QoS).

Still further, the proposed DCWSN presents an RF front-end hardware which is optimized for discrete channel switching. The cognitive engine provides a finite number of channels which can be used by the RF front-end hardware to tune it to only those channel frequencies. Hence, the spectrum sensing time is reduced and also the design level difficulties faced by VLSI engineers in facilitating channel switching as now the transmitter will only switch in between a finite number of channels as suggested by the cognitive engine.

In addition, the wireless sensor networks do not capitalize on the fact that for most applications wireless sensor networks are data centric. Further, present day cognitive radio based techniques search for free channel / white spaces continuously with respect to time, band and space which results in wastage of energy and does not take the advantage of the patterns involved in the free channels with respect to network traffic. Moreover, channel switching is done on a nearly continuous band of channels/spectrum which makes it difficult to make an RF front-end to transmit and receive at all the frequencies i.e. for a continuously large band of channels. In an embodiment of the invention, a transceiver must be reconfigured through the software, that is, radio based nodes will be required to support cognitive capabilities.

Further, to take advantage of discrete channel switching/sensing thing, a bunch of reconfigurable filter banks will be required as only a countably finite number of channels will be there to tune for or to sense for in contrast to the wideband receivers in current cognitive radios. Once the pattern of countably finite number of channels is acquired, then the reconfigurable hardware can be easily tuned for that countably finite number of channels for some time. If the pattern changes then it can be re-tuned for new countably finite number of channels. Still further, wideband receivers are very costly to make because it will require a correspondingly very fast analog to digital converter and wideband analog to digital converter are very costly. Thus, with the knowledge of countably finite number of channels in hand, the requirement of a wideband Analog to Digital converter can be relaxed.

Hence, the present disclosure utilizing the data-centric, combined localization, routing and fusion techniques, doubly cognitive architecture and optimized RF front-end hardware can not only save node energy to increase the lifetime of the network and reduce the spectrum sensing time but also ensure error free and efficient transmission.

Referring to the accompanying drawings, FIGURE 1 shows an exemplary overview of the system to monitor natural/artificial phenomenon in an area using doubly cognitive wireless sensor networks. The system comprises a plurality of reconfigurable wireless sensors nodes which are randomly distributed in an area. For instance, in case of a vehicular wireless sensor network the wireless sensor nodes are distributed and placed across highways/roads and in vehicles for detecting accidents as well as controlling traffic congestion. Figure 1 shows one such wireless sensor node with reference numeral 100.

In accordance with one aspect of this disclosure, the wireless sensor nodes 100 are arranged in a set of clusters, Figure 1 shows one such cluster represented generally by reference numeral 102. Each of these sensor nodes are adapted to elect one of the sensor nodes in the cluster as the cluster head 104 and send sensed variables to it 104. The sensed variables, for instance in case of a vehicular wireless sensor network will be velocity and acceleration of a vehicle and these variables can be used to estimate the severity of an accident. In accordance with another aspect of this disclosure, the variables sensed by the nodes are of type interval valued or fuzzy, wherein, the crisp measured values at each sensor are converted into an interval and the reasons for doing so are to achieve fault tolerance. Further, a measured crisp value is mapped into an interval by using 'left tolerance' value and 'right tolerance' values that is a reading of 2 degree centigrade temperature value lies in the interval [2° C, 3° C], whereby, at the time of fusion, intervals are fused. Further, from the sensor fusion point of view at the cluster head or base station the temperature can either be low, medium or high. Thus, these linguistic variables are represented by fuzzy sets. For instance, a temperature value of 4°C has membership 0.7 in low fuzzy set, 0.3 in medium fuzzy set and 0.1 in high fuzzy sets. In an embodiment of the invention, a collection of sensors locally form a cluster and elect a local cluster head. The cluster head receives sensed measurements from the members. Further, there are following possibilities for fusing the measurements: a. Fusion of measurements leading up to a crisp value, including:

1. Cluster head computes the means. 2. Cluster head computes the median. 3. More generally cluster head computes a symmetric function of measurements from cluster members.

b. Fusion of measurements leading to an interval valued that is using the crisp values from a sensor by allowing left and right tolerance, whereby, an interval value is attained that is 3°C measured temperature is declared as the interval [1,5]. These intervals from sensors are fused using fusion functions such as M, F, Q, N or others.

c. Fuzzy sensor fusion at the cluster head.

In accordance with still another aspect of this disclosure, the cluster heads 104 are responsible for accepting data packets from the other nodes in the cluster, aggregating them and transmitting them to a base station 106. The cluster heads 104 are also assigned the role of scanning the spectrum to determine a spectrum hole for carrying out the transmission. Thus, all the nodes in the network do not engage in spectrum sensing, which leads to significant energy savings. Also, as only the cluster heads 104 are involved in the transmission, the data packets are not flooded in all directions like in case of present day systems, instead they follow a data-centric based routing technique for reaching the base station 106.

In accordance with this disclosure, the base station 106 performs the tasks of controlling and communicating with the wireless sensor nodes 100 and receiving the data packets. Along with this the base station 106 also facilitates the creation of set of clusters by partitioning the area into a plurality of levels {L, L-1, L-2..} and sectors {S-1, S, S+l, S+2..} using incremental power broadcasts. The process of facilitating creation of clusters is described in detail, hereinafter.

The base station 106 using broadcasts based on increasing power, partitions the area to be monitored into levels (of increasing radius), wherein each level is assigned a level id. For instance, sensor nodes 100 close to the base station 106 that hear low power level broadcasts (i.e. 2mW) are labeled with level id L0. Further, all the sensor nodes 100 that hear the next powered broadcast (i.e. 5 mW) and have not set their level id, mark their level as L2. This procedure continues till the complete area is partitioned into levels. This approach may fail when the channel constitutes a fading channel.

In such a case hop- count based approach is used for leveling the area.

Still further, using a directional antenna at the base station 106 the area is further partitioned into different sectors, wherein each sector is assigned a sector id. More explicitly, using a directional antenna based broadcast the sector id is declared as 1, and all the nodes that hear the broadcast set their sector id as 1. Then, the directional antenna is steered angularly and the broadcast for sector id 2 is transmitted. The procedure continues until the entire area is divided into sectors and all the nodes 100 have their level id and sector id set. Thus, a particular node 100 in the area can be easily localized by its level id and sector id. While forwarding the sensed variables to the cluster heads 104 the sensor nodes include their level id and sector id. Therefore, in case of a vehicular wireless sensor network for instance, if an accident has occurred the authorities can easily localize the area of the accident/ vehicle by the sector id and the level id of the node that transmitted the sensed variables based data packet indicating occurrence of an accident.

Data-centric routing in accordance with one more aspect of this disclosure is performed as follows, a cluster head 104 that receives a broadcast checks whether the transmission is from a node 100 located in a higher level. If not, the cluster head 104 drops the packet. Otherwise, it broadcasts the packet. In a similar manner, the cluster heads 104 broadcast packets from the same sector or one sector above/below or a sector which is one hop away above/below from the cluster head 104. In this way, the sensed information is aggregated at the cluster head 104 and forwarded directly towards the base station 106, preventing it from propagating in all directions as in flooding. Therefore, the data-centric approach enables localization, fusion and routing operations to be performed in a combined fashion.

In accordance with a further aspect of this disclosure, the wireless sensor nodes 100 are cognitive radio based sensor nodes which facilitate the WSN to use the licensed spectrum efficiently and have an error free transmission without causing any interference to the primary users / licensed users. FIGURE 2 of the accompanying drawings shows the block diagram of a cognitive engine 200 which is embedded in a wireless sensor node 100, specifically embedded in the cluster heads 104. The cognitive engine 200 used in accordance with this disclosure, performs a dual role, firstly to determine the white space information before transmission and reception and which channels to sense to arrive at a good QoS (Quality of Service) and secondly to optimize the radio frequency parameters of the node to dynamically reconfigure the physical layer parameters to achieve maximum throughput.

The cognitive engine 200 uses a doubly cognitive module to analyze data, recognize patterns and generate training patterns for supervised learning. During the training phase, the cognitive engine 200 continuously collects information on free channel availability with respect to spatial, temporal as well as with respect to network traffic density.

The present day cognitive radio modules only determine free channel information based on spatial and temporal aspects but the doubly cognitive architecture proposed in this disclosure will not only recognize the pattern with respect to time and space but also with respect to network traffic i.e. the sensing resources will be allocated to a channel/frequency where there is less traffic apart from the preferred time and preferred space requirements. Hence, cognition with respect to time, space as well as network traffic is achieved by this disclosure.

Also, the radio frequency environmental information including fading, interference, SNR, location, visual clues on each channel is collected by the cognitive engine 200 for a sufficient long time in the area in which the DCWSN is supposed to be deployed. The cognitive engine 200 creates a long time repository (not shown in the figure) of the above information then it uses its learning unit 204 for generating training patterns for supervised learning. The cognitive engine 200 is then trained to sense for the spectrum only in the bands or channels where there is a pattern of less traffic spatially/temporally as seen in FIGURE 3, where the white blocks in the graph represent white spaces or spectrum holes which are the bands free or available for transmission.
The cognitive engine 200 continuously senses the spectrum via the sensing unit 206 based on the generated patterns stored in its learning unit 204 and the current spectrum statistics to determine the bands or channels where there is a pattern of less traffic spatially/temporally. This approach not only saves the energy resources of the nodes but also reduces the sensing time drastically. This is so because it utilizes the spatial pattern to find out which channels to sense and temporal patterns to find out when to sense those channels with an added dimension of network traffic density.

After this first stage of spectrum sensing the second stage of multi-objective optimization is done for the radio frequency environmental parameters. This multi-objective optimization is performed by the optimization unit 208 based on the learning which was achieved from the earlier optimized settings of the radio transceiver 202 and 210 done in several types of radio frequency environment situations. Thus, the cognitive engine 200 can change the radio transceiver 210 parameters based on the output of the optimization unit 208 to dynamically reconfigure the physical layer parameters to achieve maximum throughput. Hence, the wireless sensor nodes in accordance with this invention are termed as reconfigurable nodes as the nodes can dynamically reconfigure their transceivers to scan a finite number of channels and also reconfigure their RF environmental parameters to arrive at a good QoS.

In addition, the wireless sensor nodes 100 also include an optimized RF front-end unit (not shown in the figures) which dynamically tunes itself or reconfigures itself to only a countable finite number of channel frequencies in accordance with the finite number of channels determined by the cognitive engine 200. Thus, the RF front-end unit does not have to be optimized for an enormously large range of channel frequencies and this also eases the designing level difficulties faced by the VLSI engineers. The optimized RF front-end unit is implemented using Field-Programmable Gate Array (FPGA) in accordance with this invention.

In accordance with this disclosure there is provided a method for implementation of a doubly cognitive wireless sensor network for monitoring natural and artificial phenomenon in an area as seen in FIGURE 4, comprising the following steps:

• distributing a plurality of reconfigurable, wireless sensors nodes randomly in the area 1000;

• partitioning the area into a plurality of levels and sectors using incremental power broadcasts by a base station and localizing each of the wireless sensor nodes based on the level and the sector information sent in the power broadcasts 1002;

• arranging the wireless sensors nodes in a set of clusters, wherein the sensor nodes are adapted to elect and send sensed variables based data packets to a cluster head 1004;

• accepting and routing the data packets at a cluster head if the data packets are sent from a wireless node located at a higher level and in a same sector or a sector which is one hop apart above or below the cluster head 1006;

• discarding the data packets if they are not received from a wireless node located at a higher level and in a same sector or a sector which is one hop apart above or below the cluster head 1008; and

• continuously sensing spectrum availability at a cluster head for a preconfigured number of channels, aggregating and transmitting sensed variables based data packets through an available channel to the base station 1010.

Typically, the step of continuously sensing spectrum availability includes the following steps:

employing supervised learning techniques for sensing free channels for transmission and reception based on spatial, temporal and traffic related parameters and performing radio frequency environmental parameters' optimization for reconfiguring physical layer parameters for maximum throughput; and

o dynamically tuning radio frequency to a countable finite number of channels using an optimized RF front-end unit for performing channel switching.

In an embodiment of the invention, in a temperature monitoring wireless sensor network, an 'event emergency' is typically characterized by onset of fire. Further, it is essential that this event is effectively communicated to the base station as reliably and as soon as possible. Thus, event emergency leads to the following:

1. At highest available power level value, event is reported to the base station.

2. On multiple channels, the event is reported.

Further, even if sector id 1 and 2 cause poor quality of service to other wireless networks, event (say fire in WSN) reporting is given priority. In a data centric approach, base station queries the network. For instance, a typical query is all nodes having temperature above 40°C report the value. In such a network, most of the sensors are in sleep mode (at least receivers) and wake up a query of the above form is received. Furthermore, sleep and wake up approach is implemented as in case of cellular wireless phones.

TECHNICAL ADVANTAGES

The technical advantages of the present invention include providing a system for monitoring natural and artificial phenomenon using doubly cognitive wireless sensors networks.

The proposed system is an energy efficient system as it significantly saves battery of sensor nodes by employing energy aware data-centric techniques where node localization, routing and data fusion are performed in a combined fashion. The data- centric approach enables the network to collect only that part of the data which is required at a particular time or event. Also, it saves the network bandwidth and Tx/Rx time.

Moreover, the proposed system partitions the area to be monitored into clusters wherein the cluster have a cluster head which broadcasts the sensed data variables received from the wireless sensor nodes directly to the base station. Thereby, the cluster heads prevent the data packets from being propagated in all directions as in flooding and eliminate the possibility of redundant data packets reaching the base station. The proposed system also reduces the spectrum sensing time and node energy by restricting the spectrum sensing tasks to cluster heads. Thus, instead of several nodes continuously sensing the spectrum for the whole time, only a few cluster heads sense the availability of the spectrum in only a selected finite number of channels. This reduces the spectrum sensing time and increases the network lifetime.

Further, the proposed system employs machine learning techniques using a cognitive engine. The cognitive engine determines optimum radio parameters and a finite number of channels based on the sensed spatial, temporal and traffic based parameters. The optimized parameters and finite number of channels enable the cluster heads to restrict their spectrum sensing and channel switching operation to this finite number of channels, thereby, reducing spectrum sensing time and saving sensor node energy.

Still further, the present disclosure proposes an optimized reconfigurable RF front-end hardware in the sensor nodes. The reconfigurable RF front-end hardware enables the RF unit to tune its spectrum sensing and channel switching operations to the finite number of channels determined by the cognitive engine. This enables the wireless sensor nodes to opportunistically carry out error free and efficient transmissions in licensed bands and easily switch to another spectrum hole in case, presence of a PU is detected in the same band.

Thus, the present invention not only provides an intelligent data-centric routing technique but also increases the network lifetime, reduces spectrum sensing time and increases the chances of getting white spaces whenever required. The present invention can also be used in the field of Mobile Communications where similar problems of spectrum sensing time, battery life, need for co-operative and channel specific communications are faced.

Apart from that the present invention can also be used in vehicular sensor networks. In vehicular sensor networks the wireless sensor networks can be employed on vehicles and besides the highways/roads to detect accidents and control traffic to prevent congestion in the road network. The sensors will be useful in informing the traffic police, firefighting authorities and ambulance in minimum period of time, if an accident takes place. And, the fuzzy variable concept will prove useful in order to compute the accident severity.

In addition, the sensor nodes can also be used to sense velocity and acceleration for monitoring vehicles. The vehicular sensor network as well can utilize the TV spectrum bands by Cognitive Radio Technology in an opportunistic manner to overcome the issue of spectrum scarcity. Further, stereo-transceivers can be utilized for preventing accidents by embedding the stereo-transceivers into the fender of vehicles to transmit/receive electromagnetic waves. Moreover, traffic lights at an intersection can also be controlled by taking into account the number of vehicles in the lanes approaching the intersection by using the sensed information.

While considerable emphasis has been placed herein on the particular features of this invention, it will be appreciated that various modifications can be made, and that many changes can be made in the preferred embodiment without departing from the principles of the invention. These and other modifications in the nature of the invention or the preferred embodiments will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter is to be interpreted merely as illustrative of the invention and not as a limitation.

We Claim:

1. A system for monitoring natural and artificial phenomenon in an area, said system comprising:

• a plurality of reconfigurable, wireless sensors nodes randomly distributed in the area and arranged in a set of clusters, wherein said sensor nodes are adapted to elect and send sensed variables to a cluster head,

said cluster head adapted to continuously sense spectrum availability for a preconfigured number of channels and further adapted to aggregate and transmit sensed variables based data packets through an available channel using data-centric routing; and

• a base station adapted to control and communicate with said wireless sensor nodes and receive said data packets,

wherein said base station further adapted to facilitate creation of said set of clusters by partitioning the area into a plurality of levels and sectors using incremental power broadcasts and localizing each of said wireless sensor nodes in said set of clusters based on said level and said sector information sent in said power broadcasts.

2. The system as claimed in claim 1, wherein said sensed variables are selected from the group consisting of interval valued variables and fuzzy variables.

3. The system as claimed in claimed 1, wherein said wireless sensor nodes are cognitive radio wireless sensor nodes, wherein said cognitive radio wireless sensor nodes include a cognitive engine adapted to employ supervised learning techniques to perform spectrum sensing and radio frequency environmental parameters' optimization based on spatial, temporal and traffic related parameters.

4. The system as claimed in claim 1, wherein said wireless sensors nodes include an optimized RF front-end unit adapted to dynamically tune its radio frequency and transmission characteristics based on predetermined inputs.

5. The system as claimed in claim 1, wherein said cluster head is further adapted to perform data-centric routing, wherein said cluster head is adapted to accept data packets sent from a wireless sensor node and/or cluster head located in a higher level and in a same sector or in a sector which is one hop apart above or below said cluster head.

6. The system as claimed in claim 1, wherein said cluster head is still further adapted to perform data-centric routing, wherein said cluster head routes data packets received from sectors one hop apart either above or below from said cluster head.

7. The system as claimed in claim 1, wherein said data packet includes location information (level and sector) of a wireless sensor node, which sent the sensed variables in said data packet.

8. The system as claimed in claim 1, wherein the base station is adapted to use hop- count based approach to partition the area into a plurality of levels in the event that the wireless network experiences fading.

9. A method for monitoring natural and artificial phenomenon in an area, said method comprising the following steps:

• distributing a plurality of reconfigurable, wireless sensors nodes randomly in the area;

• partitioning the area into a plurality of levels and sectors using incremental power broadcasts by a base station and localizing each of said wireless sensor nodes based on said level and said sector information sent in said power broadcasts;

• arranging said wireless sensors nodes in a set of clusters, wherein said sensor nodes are adapted to elect and send sensed variables based data packets to a cluster head;

• accepting and routing said data packets at a cluster head if said data packets are sent from a wireless node located at a higher level and in a same sector or a sector which is one hop apart above or below said cluster head; and

• continuously sensing spectrum availability at a cluster head for a preconfigured number of channels, aggregating and transmitting sensed variables based data packets through an available channel to said base station.

10. The method as claimed in claim 9, wherein the step of continuously sensing spectrum availability includes the following steps:

• employing supervised learning techniques for sensing free channels for transmission and reception based on spatial, temporal and traffic related parameters and performing radio frequency environmental parameters' optimization for reconfiguring physical layer parameters; and

• dynamically tuning radio frequency to a countable finite number of channels using an optimized RF front-end unit for performing channel switching.