Communication in military environments is a highly important component during the time of war to command or share information with soldier groups, positioned in various destinations or locations. In the conventional domain, military communication is carried out manually which is not quite effective owing to delayed information delivery. MANET is the most significant technological advancement that can be used in the war fields, such that military communication can be made convenient. MANET is the infrastructure-less environment in which the routing path would be created when military communication is necessary. It will create the route path by making use of the nodes, which exist inside the coverage area. In case of sharing of different information to two troops of soldiers who are located in the same region might share the same nodes share them. In this condition, there may be a possibility of occurrence of interference issues, therefore packet loss or corruption would happen. These issues are required to be solved so as to share the information with the soldiers, such that they can direct their troops in the right manner. In this work, the issues seen in the available work such as neighborhood-based interference aware routing are solved [15][16].
The available research technique can discover the interference effect just in stable mobile node positions. But in the military war field, soldiers would have to move from one position to another dynamically and the available research work cannot see the interference effect with efficiency. In addition, because of the mobility behavior of the soldier, path disruption may happen where the information sharing will be either late or lost. These issues are solved in this research work by proposing QoS-IRDMANET. In this research work, the route is created by considering the QoS factors besides the interference limitations. Besides this, the priority of the data communication is rendered through the allocation of more bandwidth to the respective communication at execution time, in such a way that significant and critical information shall be immediately shared with the respective troop. Also, this research is concerned with the path disruption occurring because of the unavailability of resources or node movement.
-
Establishing routes when QoS parameters are satisfied
-
Identify interferences in communications for re-routing packets
-
Locate the interference effect in a dynamic manner with due consideration to the node’s mobility behavior
-
Since, the aforesaid steps are carried out for achieving an improved MANET routing, the proposed scheme can be used in military environments. The proposed framework’s flow is depicted in Fig. 1.
The proposed scheme’s objective is to achieve robust and secure distribution of information amongst military men irrespective of location. QoS-IRDMANET is detailed below in further sub-sections.
A. Establishing Routes with QoS Awareness in Military Environments
MANET wireless nodes have limitations in their energy resources. Any routing protocol that fails to consider this disability of nodes is bound to fail to result in a routing where packets might be dropped or delayed. Though this lag is permissible in a normal environment, in a militarized zone it is a great disadvantage as communications have to be swift and accurate for success or victory. This makes it imperative to consider QoS as a primary factor in route establishments. Technically this narrows down to the aspects of remaining power, bandwidth used, the stableness of the links, and end-to-end delays which need to be considered while selecting routes. Nodes satisfying the afore listed parameters are selected in route establishments. This work also uses a weighted sum in its route establishment calculations.
1. Available Bandwidth (BW): Available bandwidth (BW) specifies the available link bandwidth in the path between a source node and the destination node multicast tree. The available bandwidth is computed by using Eq. 1.
$$\text{B}\text{W}= {\alpha } {\text{B}\text{W}}_{\text{L}}+ \left(1-{\alpha }\right)\frac{{\text{T}}_{\text{i}\text{d}\text{l}\text{e}}}{{\text{T}}_{\text{p}}}{\text{B}}_{\text{c}\text{h}\text{a}\text{n}\text{n}\text{e}\text{l}}$$
1
Where, \({\alpha }\) - weight factor in the range [0, 1], \({\text{B}\text{W}}_{\text{L}}\)- obtainable local node bandwidth, \({\text{T}}_{\text{i}\text{d}\text{l}\text{e}}\)- inactive channel time, \({\text{T}}_{\text{p}}\)- time period, and \({\text{B}}_{\text{c}\text{h}\text{a}\text{n}\text{n}\text{e}\text{l}}\)- capability of a channel in bits/second.
2. Remaining Power (P): Remaining energy or power of a node as a multicast tree can be represented as Eq. (2)
$$P = {P}_{Total}– {E}_{consumed}$$
2
Where, \({P}_{Total}\)- Pre-defined total node energy and \({E}_{consumed}\) - constant in each network node where \({\text{P}}_{\text{c}\text{o}\text{n}\text{s}\text{u}\text{m}\text{e}\text{d}}\) is calculated using Eq. (3),
$${\text{P}}_{\text{c}\text{o}\text{n}\text{s}\text{u}\text{m}\text{e}\text{d}}=\frac{{\text{P}}_{\text{T}\text{h}\text{r}\text{e}\text{s}\text{h}\text{o}\text{l}\text{d}}{\text{d}}^{\text{n}}}{\text{K}}$$
3
Where, \({\text{P}}_{\text{T}\text{h}\text{r}\text{e}\text{s}\text{h}\text{o}\text{l}\text{d}}\)- Predefined power threshold, d – the distance between two nodes, n - path loss exponent, and K - predetermined constant.
3. Available Delay (D ): Delay (D) stands for the maximum delay that can occur between a source and a destination node which can be calculated using Eq. (4)
$$D = N [{d}_{trans}+ {d}_{proc}+ {d}_{prop}]$$
4
Where, N - No of links, \({d}_{proc}\)- Delays in processing packets, \({d}_{prop}\) - node propagation delay ,and \({d}_{trans}\)– delay in transmissions. \({d}_{trans}\) can be calculated using Eq. (5),
Where, N - bits count and T – transmission rate
4. Stability Of Link
This metric has been introduced in this work as a QoS parameter for assessing the stability of links. This metric accounts for a link’s received signal strength, contention, and hop counts as a part of the QoS parameter. The nodes that lie within receiving limits of a transmission is the Contention count which is computed by the periodical transmission of packets to one-hop neighbors. A sender node judges its neighbor node by counting the packet received from them while a cross-layer interaction technique assesses individual signal strengths. Any node which has a higher link stability value becomes a forwarding node.
B. Weighted Sum Method
Multi-objective executions averaging values have been found to optimize solutions efficiently. Hence, this work uses a weighted sum approach or a posteriori technique where Pareto optimal points follow the creation of the first point. A solution that satisfies pre0defined specifications is then selected. An objective function is used for optimizations and can be defined as a linear combination in presenting the weights \({\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}_{\text{j}}\ge 0\)where \(j = 1, . . . ,\beta\) (such that there is one weight for every \({f}_{j}\left(x\right)).\)Hence, the novel objective function can be expressed as below:
$$\sum _{\text{j}=1}^{{\beta }}{\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}_{\text{j}}{\text{f}}_{\text{j}}\left(\text{x}\right)$$
6
Furthermore, the weights are supposed to be normalized such as
$$\sum _{\text{j}=1}^{{\beta }}{\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}_{\text{j}}=1$$
7
Every probable value of m is treated as a bi-objective job scheduling issue in this proposed technique where the function is:
$${\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}_{1}{\delta }+{\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}_{2}\sum _{{\omega }\in {\Omega }}{\text{c}}_{{\omega }}{{\chi }}_{\text{w}}$$
8
In addition, based on Eq. (7), weight2 can be got as \(1-weigh{t}_{1}\) and hence, the objective function (8) can be modified into
$${\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}_{1}{\delta }+(1-{\text{w}\text{e}\text{i}\text{g}\text{h}\text{t}}_{1})\sum _{{\omega }\in {\Omega }}{\text{c}}_{{\omega }}{{\chi }}_{\text{w}}$$
9
Varying weight1 between the range 0 and 1, numerous prominent solutions can be found. Algorithm 1 computes the weighted sum assuming a cloud computing environment for the multi Objective job schedule issue and where \(\varLambda\)refers to a parameter fixed to 0.5.
Algorithm 1
The weighted sum based algorithm.
Input:\(\varOmega , c\omega \forall \omega \in \varOmega ,wj\forall j\in J, u,\varLambda\)
Output: The Pareto front F
Set the iteration counter\({i}_{c} = 1;\)
Set \(F: = \varnothing\);
While \(weigh{t}_{1}\le 1\) do
Solve the single-objective model and
Let x \({i}_{c}\) be the optimal solution;
Set\(F: = FU \left\{{x}^{ic}\right\};\)
Set\(weigh{t}_{1} = weigh{t}_{1} + \varLambda ;\)
Set\({i}_{c} = {i}_{c} + 1;\)
end
Eliminate dominated solutions from F;
C. Path Breakage Detection to Prevent the Packet Loss/Delay
In military communications, ruptured paths result in information delays or losses. Hence, this work has proposed a scheme to overcome path breakage with an enhanced neighbor selection method where the node’s movement outside the coverage area or its non-existence is traced. This s-twoack technique can be implemented in routing protocols as the technique creates a route from the source packet. The packets in s-twoack are called twoack packets acknowledge each other and operate similarly to the twoack scheme that routes in a predetermined 2-hop route with 3 nodes in opposite directions. A relayed packet from nodes is validated by their routing agent on successful receipt and at a 2-hop distance from the source. The success is determined on receipt of a data packet with special acknowledgments to a 2-hop distance two ack packet and the source route.
When these acknowledgments fail to reach their respective targets (sending/forwarding node) then the next hop link is assumed to be non-existent and the forwarding route is scratched. This assumption becomes the base for routing protocols to reframe routes as they ignore suspicious links in their further packet forwards, thus enhancing the network’s throughputs. Though this method of transmission used by s-two ack also results in an increased packet traffic in the network, the technique minimizes this extra traffic by waiting till a threshold number of packets accumulate in the same triplet before transmitting them in a batch. This minimizes the load as only one acknowledgment is used for the batch instead of individual acknowledgments. Disruptions in the path are identified from a list of data packet IDs maintained by the router that include nodes that are yet to receive a twoack acknowledgment packet at 2-hop distance. Nodes maintain their own list of forwarding links for their routing. Referring to Fig. 1, assume R1 does not know when R3 receives a packet, this condition is prevented in the twoack scheme as R3 sends twoack packets to the source in R2-R1 path. This updates R1 on the packets received the status of R3. When R3 is missing i.e. out of range or dead, the source receives an ERROR.
In the conditions of the node missing in the environment, the inquiry message will be transmitted to the neighboring nodes so as to find if the node is dead or has moved out of the coverage area. On the basis of the neighborhood node information and the present node information, it would be determined if that node is alive or dead. It is carried out through the analysis of the neighboring node table information for the missing node identity. In case the missing identity exists in the table along with the updated and the current seq number then it can be concluded that the neighbor node has moved out of coverage area and it is still alive. Else it can be determined that the node is dead. In this case, immediate rerouting would have happened from the node prior to the missing node rather than rerouting from the source node. By this, time complexity can be minimized significantly and the troop of soldiers can receive their information rapidly.
D. Inference Deterrence using Path Disruption/Neighborhood Information
A receiving node may receive more than one packet during transmissions creating interferences or duplications. In the case of any specific node, only its neighbors create interferences while sending/receiving data packets while other nodes are away from creating interference for the node. A path’s interference index is the same as the sum of the link index’s interferences. Hence, in a MANET’s single channel TDD (Time Division Duplex) network transmissions occur assuming the existence of only one node for transmitting information amongst the receiver’s neighbors. It is also important to note that all mobiles do not function as a sender/receiver concurrently.
E. Minimizing Interferences using Node with Lesser Neighbors
GBR-CNR scheme’s objective in design was to reduce interferences by considering nodes with less number of neighbors near the receiving node for efficiency. The underlying principle was that receiving nodes with less no of neighbors have a lesser possibility of their neighbors acting as transmitting nodes in the sender’s time slots. If A is the sender, B is the neighbor of A, and D is the destination node. Node A is likely to choose B2 for the next hop when it has lesser neighbors in comparison with B1. Improved throughputs occur in networks when receiving nodes have lesser neighbors in the neighborhood.
F. Minimizing Interferences with Lesser used Nodes
A version of GBR-CNR uses less-used nodes for minimizing interferences in transmissions called the GBR-CNR-LU. Assuming A is the senders, B is the neighbor of A and the destination is D, then two paths are available to D. B1 is the next hop destination for A1. A protocol establishing an alternative path will select A2 for its next hop and B2 will be chosen as it is less used and may be at a greater distance from D2 when compared with B1. Messages reach destinations properly while formulating paths with lesser used nodes as they are insusceptible to weakening of messages.