In the meantime, node also tries to start a sender initiated communication standard mode to by sending a pulse signal.
Publications @ Center for Distributed and Mobile Computing
However, node is deaf for in this moment, since is beamforming towards. So, node cannot hear the pulse signals sent by node , thus causing the growing of exponential backoff window of node. This situation persists until the deafness prediction mechanism of node estimates that flow may be suffering from deafness. This mean that node estimated that node has packets to send to identified as deafness by deafness prediction mechanism of node.
Therefore, enters in deafness mitigation mode, and sends a tone to node , trying to start a receiver initiated communication. When receiving the tone signal, node verifies that the destination of that signal is node , thus cancelling its backoff time and sending DATA frame to. So, the communication proceeds normally. The analytical analysis is presented below, followed by the simulation results. Both sender and receiver initiated communications are addressed in the theoretical analysis.
- 52 Things Kids Need from a Dad!
- Siuli Roy - Google Scholar Citations!
- Andrew Jackson;
The packet size considered were , , , and bytes. The improvement gets higher when short packet size and higher transmission rate are considered. In such cases, the time spent exchanging control frames consumes a higher portion of. The theoretical analysis for the receiver initiated communications is similar to that presented above. In this case, note that backoff time is not included in calculation, since the receipt of an RTR frame cancels the remaining backoff time. In this calculation, the considered values of packet size and channel transmission rate were the same used for sender initiated communications.
Note that backoff time is not included in calculation again, once the receipt of a tone-ri cancels the remaining backoff time. The packet size and channel transmission rate were the same as those in the sender initiated case. As before, the relation always holds. Similar to sender initiated communications, DPTCR-DA improvement raises when packet size is short and channel transmission rate is high. As already explained, this occurs because in these cases the time spent with RTR represents a higher portion of.
The simulation results will be presented in the next subsection. The empirical results are compared with the theoretical ones to validate the referred analysis. For this evaluation, the EXata simulator, version 2.
- Use of Smart Antennas in Ad Hoc Networks.
- Swords from the Desert.
- Sufi Bodies: Religion and Society in Medieval Islam.
- Enhancing the Performance of Ad Hoc Wireless Networks with Smart Antennas.
- Shop now and earn 2 points per $1.
The EXata simulator has a number of features, such as realistic antenna models and patterns, propagation models free space, two-ray , the IEEE It is important to recall that these simulators are widely used in related studies. Although DPTCR-DA is able to operate with other antenna types, in the simulations it was assumed the use of a switched beam antenna with 8 sectors of.
Packet size and channel transmission rate values has the same values used in the theoretical analysis. Each one of the presented results is the average of 20 simulations. Figure 5: Evaluated scenarios. For this flow, the considered packet arrival interval was defined as ms.
The packet size and the transmission data rate values considered were the same used in theoretical analysis. In the best case, the obtained throughput improvement was over higher than that experienced by the DVCS. Note that the above results are consistent with the theoretical analysis. The jitter reduces and for flows and , respectively. The correct DNAV update indicates also that the mechanism which predicts the signal source and duration works properly, as expected for the considered precisions of angle of arrival and received signal strength dB.
Recommended For You
For all the considered flows, the packet arrival interval in this scenario were , and ms. The packet size was considered to be equal to bytes, and date transmission rate was considered to be equal to Mbps. Note that this scenario may cause nodes and to be deaf to the incoming packets from node. Thus, the goal of this scenario is to show that DPTCR-DA is able to mitigate deafness problem in cases where there are alternate flows.
Beamforming in Ad Hoc Networks: MAC Design and Performance Modeling
This table also displays the fairness index values. The Jain index provides values in the interval [0,1], since it is a normalized index. Let, denote the throughput of the flow. However, in this case node has more than one flow, which causes the occurrence of alternate packets in the queue of node. Nodes and are deaf to node , thus leading to a low throughput of flows and.
Such reduction is mainly due to the time spent to mitigate deafness. Clearly, fairness and throughput are contrasting parameters. One can easily improve throughput at the expense of fairness.
As deafness cannot be handled by RI-DMAC in this case, the overall network throughput is higher at the expense of flows and. It is important to state that, in a multihop scenario, it is paramount to improve packet relay time. The empirical evaluation was also held in terms of jitter and end-to-end delay. In deafness prone scenarios, the results have shown that the DPTCR-DA was able to mitigate deafness and provide significant gains in terms of fairness, jitter, end-to-end delay and throughput of the affected flows.
The use of directional antennas has been considered a promising alternative to improve spatial division and throughput. However, to realize its benefits, efficient directional MAC protocols are needed. This work shows the benefits as well the problems associated with directional communications. Among the problems discussed, the deafness is one which has received considerable attention in the literature.
Thus, the DPTCR-DA constitutes a powerful deafness mitigation technique while providing remarkable improvements in terms of throughput, fairness, jitter, and end-to-end delay of the flows that are affected by the deafness problem. As future works, it would be interesting to evaluate different election policies, in which concerns to the node that receives a tone in the beginning of a receiver initiated communication.
Evaluation of the energy consumption gains shall be conduced as well. This work was partially supported by CNPq. Bordim, A. Barbosa, M. Caetano, and P. IEEE, , pp. Gupta and P. Kumar, V. Raghavan, and J. Mohapatra and S. Springer, Caetano, and J. Krishnamurthy and S. Springer, , vol. Choudhury, X. Yang, R. Ramanathan, and N. ACM, , pp.