• No results found

A Highly Reliable and Efficient Route Recovery Routing Algorithm for Smart Grid Neighborhood Area Networks

N/A
N/A
Protected

Academic year: 2020

Share "A Highly Reliable and Efficient Route Recovery Routing Algorithm for Smart Grid Neighborhood Area Networks"

Copied!
9
0
0

Loading.... (view fulltext now)

Full text

(1)

14th International Conference on Wireless Communications, Networking and Mobile Computing (WiCOM 2018) ISBN: 978-1-60595-578-0

A Highly Reliable and Efficient-Route-Recovery Routing Algorithm

for Smart Grid Neighborhood Area Networks

Zengwei Zheng1, Yu Zong and Meimei Huo

ABSTRACT

In Smart Grid Communication Networks (SGCNs), the Neighborhood Area Networks (NANs) locates in the middle place and transmits data up and down. The NAN routing algorithm is the key technology to determine the quality and efficiency of SGCNs communication services. Hybrid wireless mesh protocol (HWMP) is available for NAN, but not perfect. This paper explores the route recovery mechanism of HWMP and put up some improvements to it. Combined with our research on routing metric of HWMP, a new improved routing algorithm HR-HWMP is proposed in this paper. NS3 is used to simulate the performance of HR-HWMP. The results show that the performance in the packet transmission rate and network throughput has been improved.

1 INTRODUCTION

With the rapid development of the internet of things (IoT), IoT has been significantly applied in our daily lives, industry and farming. In our daily lives, some challenges arise in the traditional electricity networks, including the user's demand response and optimal allocation of energy saving equipment. The traditional electricity system needs to be smart[1,2]. The next electricity power is named smart grids which is the smart power system that integrates two-way communications, sensing, and control technologies into the traditional power system [2].

NANs undertakedata transmission tasksamong SGCN, and routing algorithm plays an important role in NANs. Wireless mesh network (WMN) topology is adapted to NANs considering the communication characteristics between nodes.Many researchers do a lot of work on routing problem of NANs [3, 4].HWMP is the default routing protocol of IEEE802.11s for WMN and therefore can be adapted to NANs.

There are many factors impacting performance of a routing algorithm, such as routing metric and route recovery mechanism. Routing metric, that is the standard to choose next hop, is studied in another research work of ours [5]. This paper focuses on the route recovery mechanism of HWMP in NANs. Combined research on routing metric and route recovery mechanism, a new routing algorithm, highly reliable and efficient-route-recovery routing algorithm, called HR-HWMP, is proposed in this paper.Some simulations have been done to evaluate the performance of HR-HWMP using NS-3.The simulationresults show that the performance in the packet transmission rate and network throughput has improved. When the number of nodes is greater than 40, the packet transmission rate is about 95%.

1

(2)

The rest of the paper is organized as follows. In Section 2,we introduce the route recovery mechanism of HWMP protocol in detail. In section 3, anew route recovery mechanism is proposed to increase the performance of HWMP in NANs. Simulation results are presented in Section 4 and concluding remarks are drawn in Section 5.

2 BACKGROUND

[image:2.612.208.378.335.422.2]

As shown in Figure 1, it is assumed that the source node A will transmit packet to the destination node E. The optimal route A → B → C → E is established by path discovery for transmission. If the link between node C and node E break which will trigger route recovery mechanism of HWMP. The node C that failed to forward the packet will generate a PERR message and send it to its predecessor node of E. PERR message is generated to inform all predecessor nodes that path is not available. All of the lists, including all nodes on this link, will be maintained. After the node B received the PERR, it will check the routing table according to the PERR. And the PERR will be forwarded to the predecessor node A. When the PERR is received, the route destined to E is marked as unavailable. If node A still needs to send the packet to node E, node A will reestablish a link to node E by path discovery to transmit the packets.

Figure 1. Route recovery example of HWMP.

If the packet forwarding fails, the route recovery mechanism of HWMP is triggered, PERR is generated and sent to the predecessor node, and the new link to the destination node is reestablished by path discovery.

According to the route recovery process of HWMP, generating PERR obviously will increase the load of the communication network and reduce the throughput of the network, especially, if interruption of communication links is due to interference or other factors and when it is no need to trigger route recovery mechanism. Although there are some research work on route recovery question [6-9], no literature on recovery mechanism of HWMP is found. This paper proposed an improved route recovery mechanism for HWMP including listening mechanism and backup path recovery mechanism which will be described in detail in the following section.

3 RELIABLE ROUTE RECOVERY MECHANISM

(3)

backtracking algorithm. The recovery mechanism utilizes the backup path maintained by the node to transmit the packet in the case of the communication link breakage, and reduces the unnecessary trigger of sending PERR message. The new route recovery mechanism will be described in detail.

Listening mechanism

In a network communication, the communication link fluctuation resulting by interference or other factors causes the route recovery mechanism to happen. Broadcast PERR message can increase the network load, therefore make the network communication quality decline. Actually, in some cases, route recovery is unnecessary [7].Frequent triggers of route recovery mechanism will affect the communication quality of the network apparently. Before the route recovery process, attaining the link quality and justifying the necessity of route recovery, called listening mechanism in this paper, will reduce overhead of the whole network.

The link quality parameter Quality(i,j) can be calculated by SINR parameter. The formula is as follow[10]:

SINR = 𝑃0

𝑃𝑁+∑𝑛𝑟−1𝑃𝑟 (1)

Where 𝑃0 represents the signal strength of the receiving node receiving packet, 𝑃𝑁 is the noise level, and ∑𝑛𝑟−1𝑃𝑟 represents the sum of the noisy signals.

The link quality judgment flow between node i and node j is shown in the following Algorithm 1.

TABLE I. LINKBREAKDECISION ALGORITHM. Algorithm 1: LinkBreakDecision

Input: Quallity(i,j), link(i,j)

Output: Link Break Decision for link(i,j) 1: begin

2: BreakStatus(i,j) ← false

3: if Quality(i,j) <SINRmin then

4: BreakStatus(i,j) ← true 5: end

6: return BreakStatus(i,j) 7: end

Backup path recovery mechanism

When the communication link is interrupted and the feedback result of the listening mechanism is poor link quality, it is regarded as a broken link, and the backup path recovery mechanism based on the backtracking algorithm is enabled.

The recovery mechanism involved backtracking algorithm, backup path mechanism and threshold control technology.

(4)

The backup path mechanism mainly makes up for the flaws of HWMP that each node only maintains the optimal path. With the backup path mechanism, the packet can be forwarded by backup path when the optimal link is broken.

[image:4.612.81.517.186.427.2]

The threshold control is used to discard the backup path and back to upstream node for link selection. Comparing routing metric of the backup path and optimal path, if greater than predefined threshold N, the backup path is discarded and back to the upstream hop to select another link. Otherwise, the backup path is used to forward the packet. The backup path recovery mechanism is described as following Algorithm 2.

TABLE II. BACKTRACKRECOVERYALGORITHM. Algorithm 2: BackTrackRecovery

1: begin

2: if i has reserve route to j &&

|Pathreserve(i,j).metric - Pathcurr(i,j)|≤ N then 3: Select Pathreserve(i,j) as Pathcurr(i,j) 4: return

5: else

6:InfNei(i, Pathcurr(i,j)) 7: while (i != s)

8: if PreNode(i) has a reserve route to j &&

Pathreserver(PreNode(i),j).metric-Pathcurr(PreNode(i),j)≤N then 9: Select Pathreserver(PreNode(i),j) asPathcurr(PreNode(i),j) PreNode(i) retransmit data with Pathcurr(PreNode(i),j)

10: end 11: end

12: set PreNode(i) as i

13: s start on-demand routing to find Path(i,j) 14: end

15: end

Efficient route recovery routing algorithm

[image:4.612.99.496.550.711.2]

The new proposed efficient route recovery routing algorithm is described as algorithm 3. The listening mechanism is used to obtain the Quality(i,j). If there is a link failed, then a link break decision will be made. If the link is broken, the backup path recovery mechanism based on the backtracking algorithm will be enabled.

TABLE III. BACKTRACKRECOVERY ALGORITHM.

Algorithm 3: Efficient route recovery routing algorithm 1: begin

2: while Outgoing links at node i != null do 3:foreach link(i,j) do

4: Quality(i,j) ←LinkQualityEstmation(link(i,j)) 5: end

6: if LinkFailureTure on link(i,j) then

7: if LinkBreakDecision(Quality(i,j),link(i,j)) then 8:BackTrackRecovery(link(i,j))

(5)

For example, in the network topology shown in Figure 2, node s wants to send packets to node d. Here, node s is treated as the source node and node d is the destination node. First, the source node s selects an optimal path to the destination node d, dghijd.If node i fails to transmit data to node j, listening mechanism is used to obtain the link quality parameter Quality(i, j). Comparing the Quality(i, j)

with the predefined𝑆𝐼𝑁𝑅𝑚𝑖𝑛, if Quality(i, j) is greater than or equal to the 𝑆𝐼𝑁𝑅𝑚𝑖𝑛, it is assumed that the link from node i to node j is only temporary broken and the link from node i to node j can still complete the communication and the packet will be retransmitted by the same link from node i to node j. And if Quality(i, j) is less than 𝑆𝐼𝑁𝑅𝑚𝑖𝑛, the link i to j is really broken.

Node i will check the alternate path to the destination node according to the route recovery mechanism, and if there is not, the node i will send PERR to it’s the parent node h. The node h will check the alternate path, if there is no path to the destination node d, the PERR will be sent to the its parent node g until it goes back to the root nodes. If node i has a backup pathinopqtd as shown in figure 3, the routing metric of the path will be compared with the routing metric of the optimal path. If the difference is less than or equal to predefined N (constant), the alternate path

inopqtdwill be used to complete packet transmission. If the difference is greater than N, node i will send a PERR to its parent node and inform the node h that link i to j broken.

[image:5.612.103.504.356.562.2]

(a) (b) Figure 2. An example of broken link.

Figure 3. An example of reserve path.

(6)

back to the source nodes. The source node will reestablished a route by a path discovery and complete the packet transmission with the new path.

4 HR-HWMP

In another research work on HWMP [5], a new routing metric, Hybrid Metric is proposed to improve the reliability of HWMP. Due to the limit of paper length, Hybrid Metric cannot be detailed here. Combined Hybrid Metric and the new efficient route recovery routing algorithm presented here, an improved HWMP is named HR-HWMP which means High Reliable-HWMP.

5SIMULATION

The simulator NS3 is used to analyze the performance of HR-HWMP in NANs and simulation parameters is shown in Table IV.

TABLEIV.SIMULATIONPARAMETERS.

Parameters Values

Wifi Standard 802.11a

Interface 2

Total Time 100s

Numbers of Nodes 2×8~2×20

Type of Packet UDP

Packet Size 1024

Tx power 16dBm

MeshHWMPpreqMinInterval 0.1024ns

MaxQueueSize 255

Channel Number 5

Root number 1

Step 50

PacketInterval 0.1s

The simulation were on different topologies of 2×8, 2×9,……, 2×20. The step between nodes is 50m. WiFi standard is 802.11a. The packet size is set to 1024 bytes, and the packet interval is 0.1s. The total simulation time is 100s. The detailed parameters are shown in Table IV.

Simulation results analysis

In order to evaluate the performance of HR-HWMP routing protocol, the following performance metrics are considered:packet delivery fragment (PDF), average network throughput, average end-to-end delay and the number of control message packets [11].

1) Packet delivery fragment

(7)

PDF =𝑅𝑥𝑇𝑥× 100%(2) Where Rx denotes the number of received packets and Tx denotes the number of transmission packets.

2) Average delay

The average delay is calculated as:

Average delay =𝑑𝑒𝑙𝑎𝑦𝑆𝑢𝑚𝑅𝑥 (3) Where delaySum represents the sum of all end-to-end delays for all received packets containing the data stream, and Rx represents the number of received packets.

3) Throughput

Throughput is expressed as the network data transmission capacity.

[image:7.612.176.421.232.544.2]

All simulation results of HR-HWMP are compared with raw HWMP and RE-HWMP [7].

Figure 4. PDF.

Figure 5. Network throughput.

From Figure 4 and Figure 5,it can be seen that HR-HWMP has greatly improved in the aspects of PDF and throughput of the HR-HWMP compared with the HWMP and RE-HWMP [7]. When the number of nodes increases, the difference is more obvious. When the number of nodes is bigger than 40, the packet delivery fragment can still be about 95%.

Figure 6 shows that HR-HWMP works almost as same as HWMP and RE-HWMP on average end-to-end delay.HR-RE-HWMP does not perform worse than the other two.

(8)

propose in this paper works. When the number of nodes is less than or equal to 24, the number of PERR message does not decrease obviously compared with HWMP and RE-HWMP. When the number of nodes is bigger than 24, it can be seen that the number of PERR message in the HR-HWMP protocol is significantly reduced.

[image:8.612.168.425.125.475.2]

Figure 6. End-to-end delay.

Figure 7. Control messages.

6 CONCLUSION

(9)

7 ACKNOWLEDGEMENT

This work is supported by the Zhejiang Province Natural Science Foundation of China under Grant LY15F020023, and Hangzhou Key Laboratory for IoT Technology

&Application, Zhejiang, China.

REFERENCES

1. G. Rajalingham, Q. D. Ho and T. Le-Ngoc, "Evaluation of an efficient Smart Grid communication system at the neighbor area level," 2014 IEEE 11th Consumer Communications and Networking Conference (CCNC), Las Vegas, NV, 2014, pp. 426-431.

2. W. Meng, R. Ma and H. H. Chen, "Smart grid neighborhood area networks: a survey," in IEEE Network, vol. 28, no. 1, pp. 24-32, January-February 2014.

3. Saputro, Nico, K. Akkaya, and S. Uludag. "A survey of routing protocols for smart grid communications." Computer Networks 56.11(2012):2742-2771.

4. Q. D. Ho, Y. Gao, G. Rajalingham and T. Le-Ngoc, "Performance and applicability of candidate routing protocols for smart grid's wireless mesh neighbor area networks," 2014 IEEE International Conference on Communications (ICC), Sydney, NSW, 2014, pp. 3682-3687.

5. Y. Zong, Z. Zheng and M. Huo, "Improving the reliability of HWMP for smart grid neighborhood area networks," 2016 International Conference on Smart Grid and Clean Energy Technologies (ICSGCE), Chengdu, 2016, pp. 24-30.

6. Yu, C. W., T. K. Wu, and R. H. Cheng. "A low overhead dynamic route repairing mechanism for mobile ad hoc networks." Computer Communications 30.5(2007):1152-1163.

7. Ashraf, Usman, S. Abdellatif, and G. Juanole. "Route Maintenance in IEEE 802.11 wireless mesh networks." Computer Communications 34.13(2011):1604-1621.

8. Maleki, Ahmad, and M. Sadeghzadeh. "Improved AODV route recovery in mobile ad-hoc networks using a genetic algorithm." Management Science Letters 4.9(2014):2171-2176.

9. S. A. Alvi and A. N. Mian, "On route maintenance and recovery mechanism of RPL," 2017 13th International Wireless Communications and Mobile Computing Conference (IWCMC), Valencia, 2017, pp. 1933-1938.

10. Vlavianos, Angelos, et al. "Assessing link quality in IEEE 802.11 Wireless Networks: Which is the right metric?." IEEE, International Symposium on Personal, Indoor and Mobile Radio Communications IEEE, 2010:1-6.

Figure

Figure 1. Route recovery example of HWMP.
TABLE II. BACKTRACKRECOVERYALGORITHM. Algorithm 2: BackTrackRecovery
Figure 3. An example of reserve path.
Figure 4. PDF.
+2

References

Related documents

(1) a novel iterative refinement scheme between the depth estimation and the data fusion, (2) a novel anisotropic kernel density estimation based outlier removal algorithm, (3) a

One-locus models lead to different equilibrium values because of constraints on the relationship between the mean and variance imposed by the assumptions of

The program has its rationale in cognitive development theory, research on approaches to learning and the literature concerning adult and teacher development in the

So when the number of steps is minimized (without it affecting the integral closure) we consider the generating set to be as simple as possible, and so the ideal generated by it is

We present a constructive method for the study of observability or partial observability of weakly coupled linear distributed systems or, more generally, of compactly

Further, the performance of supply chain helps to improve business efficiency and value at every stage which includes the strategic, tactical, and operational

Examples of services for which you must contact your case manager to verify coverage are home health aide, personal emergency response system, chore, home adaptation, assisted