κ-Coverage Reliability for Wireless Multihop Network incorporating Boundary Effect

Coverage is one of the crucial performance measures for Wireless Multihop Networks (WMN) and is widely explored by researchers. The coverage performance of the network indicates how effectively the deployed network is able to provide the required coverage for a given period of time. The coverage performance of a finite network is influenced by different factors such as the presence of obstacles in the signal propagation path, the shape and size of the network deployment region, sensor node characteristics, and many others. Therefore, it is necessary to compute the performance of the network before its actual deployment in the desired region. The existing paper proposes an analytical solution for the network coverage considering sensor node failure and boundary effect using three different sensing models i.e., Boolean, Shadow Fading, and Elfes, separately. The Boolean Sensing Model (BSM) is deterministic and ignores the channel randomness, whereas the other two are probabilistic sensing models and are more realistic as they also consider the channel randomness caused by environmental factors. In the existing research, coverage estimation is determined separately for each individual model. However, coverage estimation for all three models and coverage reliability under boundary conditions have not been explored in the literature yet. Coverage reliability is another crucial metric determining how long a network can provide the required coverage. This paper aims to compute the Network Coverage Reliability (NCR) of a WMN deployed in given environments and with sensor node failure conditions. We have proposed an analytical solution to estimate the NCR of a WMN deployed in a finite circular region considering boundary effect using all three models viz., BSM, Shadowing & Multipath Fading (SMF), and Elfes Sensing Model (ESM) assuming sensor node failure condition. The Mean Time to Failure (MTTF) parameter has been determined for all three models and based on that the value of NCR is estimated. The proposed work considers κ-coverage reliability with 1≤κ≤3 where κ is the number of distinct sensor node required to cover the target and render desired NCR for different path fading and non-fading sensing models. The analytical results are validated through simulation runs and found to be consistent. Further, we have analyzed the impact of different parameters, such as the number of sensor node, sensing range, SMF parameters, sensor failure rate, and required reliability on the κ-coverage and κ-coverage reliability, referred to as NCR.