An Integrated Signal Allocation Model with Effective Collision Resolution Model for Performance Enhancement of Wireless Sensor Networks

Small and affordable strategies that include a microcontroller with low power, multiple sensors and a communications radio consist of WSNs. They are independent ad-hoc networks which can sensor, process data and transmit various physical parameters to a user through the use of multi-hop communications [1]. They offer a versatile, independent, low-cost solution, particularly for places with limited accessibility, for a variety of distributed monitoring applications [2]. Given the difficulty of replacing batteries, WSNs must use efficient energy storage schemes to save energy for longer life. The use of renewable energies, such as sunlight, vibration, heat etc. for long-term operations at WSNs is an alternative solution [3].

In WSN, collision resolution is a key concern. There are disadvantages in terms of energy consumption and latency method in current collision resolution approaches [4]. If several nodes intend to concurrently send packets to the common medium, a collision occurs [5]. Unforeseen collisions harm the use of wireless signals by wasting resources and degradation of network performance. low-duty cycles are used by most WSNs as energy is reduced [6]. In order to save energy, sensor nodes try to stop their radio, which squeezes the time available for packet transmits between nodes and may lead to collisions. For event detection, for example, WSNs add a wider risk of collisions to predictable busy traffic triggered by events [7].

There have been suggestions for dealing with WSN collisions on various protocols of MAC (Medium Access Control). In principle, all of these ideas adopt the CSMA/CA mechanism. A node senses the signal in case of a collision before transmission and retrieving it randomly [8]. The main principle underlying CSMA/CA is that a station ought to be able to obtain while broadcasting in order to identify a collision between two stations. In wired networks, if a collision happens, the energy of the received signal nearly doubles, and the station detects the probability of a collision. CSMA/CA is a networking multiple access strategy wherein the nodes seek to prevent collisions by initiating transmission only once the channel is felt to be "idle."

The duration of the random back-off window is raised exponentially to prevent secondary collision [9]. The investigation demonstrates that in general wireless networks, CSMA-based techniques can produce acceptable outcomes. However, the efficiency of duty cycling WSNs may decline significantly, particularly when the network operates with uncoordinated burst traffic [10]. Node coordination is an effective method for avoiding recurring collisions [11]. RI-MAC, for example, is a WSN MAC protocol initiated by the receiver. If there is a collision, the recipient requests several senders with an exponentially increasing back off window [12]. However, this results in significant lag in distribution and energy use. As a result, the process of identifying and resolving collisions takes time but consumes no energy because only one sender is located and access to the signal is gained in each round [13]. Control packets must also be longer than usual to resolve collisions and can conflict with other packet transmissions that restrict network output further. The Network Architecture is shown in Figure 1.

1.png

Figure 1. Network architecture

Self-organization has been one of the most significant distributed strategies in recent years [14]. This method allows a sensor network to produce emergent behaviour in which nodes communicate independently and collaborate autonomously. The goal is to complete tasks that are beyond the capacity of a single node [15]. Protocols designed for dispersed sensor network must be able to provide optimal energy usage while taking into account mobile nodes, external noise, limited batteries, and message loss, among other factors [16].

In communication systems, wireless networking plays an important role. Modern wireless multimedia applications with restricted spectrum resources need a greater rate of data and higher bandwidth for multiple users [17]. In order to satisfy the portability requirements of future wireless systems, low energy use and smaller Silicone areas are crucial. The energy-efficient implementation and cost-efficiency of advanced communication algorithms are still the design challenge, requiring optimisation of wireless device design; system/algorithm, architecture and technology deployment [18].

New developments have proven or demonstrated unparalleled wireless spectral efficiency in knowledge theoretical and communication algorithms [19]. However, many of them need very high processing power and have therefore not been able to incorporate data rates in real time and using traditional devices with acceptable energy consumption and expenses for typical portable applications requiring wireless communication [20]. The distinctive features of this application domain can be exploited in order to reduce the cost of implementation. The focus of the work is on providing the wireless networks, especially wireless sensor networks with energy efficient and collision prevention protocols [21].

The networking group has made many efforts to develop routing protocols for energy saving in one-signal sensor networks [22]. Regrettably, any node suffers from overheard transmissions from all other nodes in its scope, resulting in high energy wastage when only one signal is used [23]. This can be alleviated by the use of several network signals. Using multiple signals also helps to reduce interference and increase communication efficiency in the network [24]. Current WSN hardware, like MICAz and Telos, that use the radio CC2420, provide multi-signal systems (16 signals with 5 MHz of intermediate frequency spacing). Problems imposed by energy supply fluctuations can be solved using methods to adjust their energy consumption dynamically on the basis of projected energy supplies. It is proved that selection and routing of complex signals can solve this issue [25]. The design of successful frameworks for jointly selecting energy management signals and routes is therefore a complex problem since it is necessary to resolve wide-ranging network adaptations, rather than individual node modifications.

Collisions on wireless systems can generally be a key cause of amplified latency and retransmission of packets. As energy-constrained wireless networks collide, additional latency and retransmissions are equivalent to excess power consumption. The energy on the remote platform is premium [26], so collision prevention will improve the network's life span. Current network sensor solutions attempt either with the help of TDMA or RTS/CTS to solve the collision problem in the MAC layer, but applications-specific information can't easily be exploited by MAC layer solutions [27] or they try. The collision in the network is indicated in Figure 2.

2.jpg

Figure 2. Collision model

When two or even more nodes attempt to send a packet all over the network at the same time, they collide. Typically, a node in a WSN does not know when to wake up in order to receive a data packet, thus it must maintain its radio on, which consumes the majority of the energy. A collision avoidance mechanism using domain-specific information will complement and increase the efficiency of the MAC layer, given a specific type of application [28]. When implementing a collision prevention system, wireless communication can be partitioned into two types: a single and many packages. Previous packet arrivals don't have information on potential arrivals for a single packet communication; arrivals are basically random. The presence of a 'next' packet is specified in multi packet communication. A correlation now exists between past and future arrivals. In this paper, an Integrated Signal Allocation Model with Effective Collision Resolution Model in Routing for Performance Enhancement of Wireless Sensor Networks is proposed that improves the performance of the system.

The use of Integrated Labelling system is a simple method for reducing collisions. Time is divided into vast eras, where each source selects a time to submit by random measure. Since the time is large enough relative to the time of sending packet, the likelihood of conflicting sending times for some other source is small. Per source can only send one packet per time by definition. Latency is also directly proportional to the size of the time. Ideally, the risk of potential collisions can be reduced without increasing the latency or number of retransmissions substantially [29].

Avoidance of congestion on a network for packet switching is equivalent to a collision in a wireless network. In each case, in presence of an anomaly, the source throws the data rate. The effusive rate would influence the data packets arrive and therefore avoid repeated collisions due to overlap periods in the prevention of collisions. By slowing down the source, large traffic on the receiver could also dissipate. This latter result implicitly implies that all traffic in sources is finite and that the background sources will finally stop after some time. The specification implicitly assumes that a packet loss is a result of congestion in the case of the TCP congestion control. The data from collisions or a link failure may be lost on the user This presumption cannot be found in wireless systems.

If we suppose that all sources send packets with the same rate of 1/T—that is, every active source sends a packet during time frame T, then the recipient can also track packet arrivals and detect significant periods of inactivity within a certain time frame. If a packet loss is caused by a collision, the recipient will send the "largest silent time" information to their primary source. At that time, the source could try to transmit to minimise the likelihood of further crises. The source basically tries to interlink its transmissions with background transmissions while still maintaining a steady rate of data. This technique is known as the avoidance of phase offset collisions using integrated signal allocation model that reduces the collisions.

Colliding on the recipient is not as simple as testing for corrupt packages since both connection losses and collisions can lead to corruption. We can't always inspect the packet to decide its source because it's corrupt. The only thing that can be deduced is that there was a failure. Ideally, when each source packets arrived, the recipient would know exactly. If several packages arrived simultaneously, the receiver could decide whether a collision happened and notify the source of the collision. By inserting a little determinism into the protocol, the optimal solution can be estimated. Since packets may come within the range at any time, a collision cannot occur even when ranges are overlapping. The model for collision detection cannot therefore be used to forecast possible collisions accurately. In accordance with this, if the collision occurred by the end of the scheduled arrival date, along with the transfer shift, a missing packet is defined as collision-induced loss.

The Integrated signal allocation model verifies the available signals in full length and selects the best signal and allocates for completing the data transmission. The signal strength is estimated as

$\begin{aligned} \text { Strength }_{C H} &=\left(\frac{L i}{I}-1\right) * S i+* D L_{A}+g^{*} E_{s y n c}+\varepsilon f_{s} \end{aligned}$          (1)

Here Li is the limit of the signals range, DL is the Data loss limit of signal transmitted by antenna A, Si is the signal range, E_sync is the synchronization levels of the signal, εf is the region range of the nodes. The squared signal distance level is calculated as

$E\left[d_{t o C H}^{2}\right]=\iint\left((m-x)^{2}+(n-y)^{2}\right) p(x, y) d x d y$          (2)

Here m and n are arbitrary signal region in the limit, x and y are base station location coordinates.

The estimation of Signal Frequency Sf and Time Interval gap Tg can be obtained by jointly maximizing over Sf and Tg. The operation is performed as

$C h_{\text {freq }}(S f, T g)=\frac{1}{D_{R} \times L} \sum_{i=1}^{L}\left\|m(I)-x^{*} y(I)\right\|^{T}$             (3)

Here D is the squared signal distance, T is the Thresholds range and L is limit of the signals to be allocated.

The estimate of best signal data collected from the estimated signals without loss is performed as

$D c(m, n)=\arg \left\{\min _{x, y} L_{c h}(x, y)\right\}=\iint \frac{(m-x)^{2}+(n-y)^{2}}{\operatorname{range}(A)+T h} d x d y$              (4)

The non-overlapping signal frequencies are calculated as

$C h_{f}=\frac{\sqrt{N L}}{\sqrt{2 \pi}} \sqrt{\frac{\varepsilon f_{s}}{\varepsilon_{m n}}} \frac{L}{d_{t o C H}^{2}}$.            (5)

Here

$\varepsilon_{m n}=\frac{N L_{x y}\left(N_{s r c}-B_{m}\right)+\ell_{t}\left(2^{2 W}-1\right)}{\ell_{j}\left(N L_{x y}+N_{y x}\right)+T h}+\sum_{i=1}^{L}\|m(I)-x * y(I)\|^{T}$

$B_{m}= \begin{cases}\frac{\left(2^{2 W}-1\right)}{N L_{x m}}, & B_{m}<\frac{\left(2^{2 W}-1\right)}{N L m x} \\ \tilde{P}_{m}, & \frac{\left(2^{2 W}-1\right)}{N_{z a}} \leq B_{m} \leq \frac{\left(2^{2 W}-1\right)}{N L_{n y}} \\ \frac{\left(2^{2 W}-1\right)}{N L_{n y}}, & B_{m}<\frac{\left(2^{2 W}-1\right)}{N L_{x y}}\end{cases}$                (6)

Here NL is the noise values, 2W is the bandwidth limit, Th is the threshold, m, n are arbitrary signal region in the limit, x and y are base station location coordinates. The collisions in the network are identified as

$=\frac{1}{N L} \frac{\sum_{n=1}^{N} \min \left\{C h_{f r e q}\left\{\left(\varepsilon_{n}^{L}\right)^{M} V_{n}^{L} \psi_{n}^{T_{n}}\right\}\right\}}{\sum_{n=1}^{N} \sqrt{\frac{\varepsilon f_{s}}{\varepsilon_{m n}}} \frac{L}{d_{t o C H}^{2}}}$         (7)

where, ε represents the optimal signals from the available signals.

The collision rate is reduced in the Integrated Signal Allocation Model with Effective Collision Resolution Model is performed as

$\min \left\{C h_{\text {freq }}\left\{\left(\varepsilon_{n}^{L}\right)^{M} V_{n}^{L} \psi_{n}^{T h}\right\}\right\}=C h_{f r e q}\left\{\begin{array}{l}\left(\varepsilon_{n}^{L}\right)^{M}+\max \left\{\left(\operatorname{diag}\left(\hat{x}_{n}^{L}\right)\right)^{M} * C h_{f}\left(N L_{n}\right)^{T h}\right. \\ \left.* \operatorname{diag}\left(x_{m}^{L}\right)\right\}+\left(\varepsilon_{n}^{L}\right)\end{array}\right\}$                (8)

The Normalized Mean Square Error (NMSE) of the collisions based on allotted signals are calculated as

$N M S E=\frac{\varepsilon^{2}}{C h_{\text {frea }}} \frac{\sum_{i=1}^{L} \text { Strength }_{C H}\left\{\left(\varepsilon_{n}^{L}\right)^{T h}\right\}}{\sum_{i=1}^{N} B_{m}\left\{\left\|S_{n}+D L\right\|^{2}\right\}}+\frac{N L_{x y}\left(N_{s r c}-B_{m}\right)+\ell_{t}\left(2^{2 W}-1\right)}{\operatorname{Sn}\left(N L_{x y}+N_{y x}\right)+T h}$            (9)