Designing a Wireless Sensor Network strategy which monitors and reports the air, water, data and sound quality of a given urban area of 1 km2.
Our sensor network strategy is designed keeping in mind some constraints i.e. increasing lifetime of network, low energy consumption and large area coverage.
During investigation, we examined many algorithms, noted their pros and cons, the pros of previous strategies have been implemented and minimized their cons as far as possible. This is one way which makes our design novel.
Our WSN follows cluster topology with dynamic multi-hop transmission. After cluster formation, data collection and transmission proceeds.
For cluster formation, in round 1 the nearest sensor node (SN) from base station (BS) with maximum cluster members (CM) covered, will become the cluster head (CH). Further rounds will initiate when CH's energy level goes below set threshold, the next CH will be selected based on maximum energy available among all CMs, minimum distance from BS and maximum CMs covered.
For transmission from CM to CH, multi-hopping is implemented where shortest path is covered from CM to CH, also the path goes through the neighbor CM with maximum energy left among all suitable candidates. Aforementioned process repeats for neighbor CMs for further transmission to CH.
For transmission from CH to BS, multi-hopping is implemented where shortest path is covered from CH to BS passing through other CHs which are close-by and have required energy left for transmission. To save transmission overheads, the CMs including the BS under their transmission range transmit data directly to BS.
To place the SNs across metropolis, sensing range of all sensors used to monitor air, data and sound quality have to be noted and their average will be used to decide the ultimate sensing range of SN. This sensor range is used to position sensors such that all sensors should cover maximum area, minimizing overlap. The nodes for water sensors shall be placed separately, most preferably in public drinking water taps using multiple SNs.
The node can be mounted atop existing utility poles keeping in mind the sensing range such that it covers ground and air alike and providing the necessary height to SNs for minimizing transmission losses that occur due to reflection, refraction, diffraction and attenuation caused by surroundings.
Hence 2 types of nodes are going to be made. Type 1 for air, data and sound quality and Type 2 for water quality. Both types of SNs follow the aforementioned algorithm.
Air quality is reported by measuring standardized parameters i.e. NOx, ground level ozone, PM 2.5, PM 10, CO2, SO2. Sound quality is measured using noise pollution sensor. Data quality can be measured using a Wi-Fi interface to connect to a nearby public Wi-Fi and fetch service quality details such as internet speed, number of users and total bandwidth. Water quality is reported by standardized parameters i.e. temperature, pH, dissolved oxygen, total dissolved solids, turbidity, conductivity, sediments and salinity.
The collected data from monitoring will be processed in BS, providing information from data and sent to concerned authority, which helps authorities to take better decisions and alert them of possible mishaps, if any.
Clustering algorithm has been used to reduce power consumption for transmission, because the farther the nodes are away from BS, the more power will be consumed for transmission. Multi-hopping from CH to BS has been used to reduce power consumption in transmission from CH to BS, because the farther away from the BS the CH is, the more the power is consumed. The path from CM to CH employs multi-hopping for the same reason. The CMs nearest to BS transmitting their data directly to BS to reduce power consumption as they follow the shortest path. CHs are dynamically assigned to allocate and de-allocate CH's position based on energy level remaining in node, as soon as the CH's energy level falls below a threshold, the next member who has most energy remaining and satisfying the criteria of CH will be made CH. Dynamic allocation is done to increase lifetime of network by increasing lifetime of nodes.
In multi-hopping approach, CM sends data through neighboring CM under the following three conditions- first, the destination CMs should be nearest to the source CM; second, the destination CM should have the necessary energy to do the transmission; third, it should be near to BS. Destination CMs best satisfying the above conditions will be fit for transmission. This approach saves energy of the nodes by eliminating any unnecessary transmissions in the route.
To save power between transmissions, sensor nodes are duty cycled with predefined sleep time.
For transmission media, any technology maybe used, however Zigbee is recommended for short transmissions (CM - CH, CM - CM and CM - BS) and LoRaWAN is recommended for long transmissions (CH to CH, CH to BS).
The potential impact of our design is saving power by being more energy-efficient compared to traditional algorithms. It increases the lifetime of WSN which results in a reliable, scalable and energy-efficient network, is practical to implement and hence, puts a great impact on the globe.