By planning the radio mesh network before rolling out smart meters you can save time and reduce the number of repeated visits to house owners.
As described in the blog article Work order system for effective rollout of AMS we have described how to find the best alternative position of the concentrator which we refer to as a master node.
In this article we'll focus on planning the optimal order for rolling out the smart meters in an AMR system based on the radio plan described in The best way to plan RF mesh networks.
Let’s assume we've found and possibly installed the required masters to cover the area of interest. Now the routers, repeaters and smart meters, called slave nodes, need to be installed.
Some might say that the order of installation isn't important, since the mesh network is autonomous or that the order is given by road axis and access to the planned houses.Most roads are bidirectional and it might be intuitive to think that it's irrelevant whether the order of the rollout goes from A to B rather than B to A. But as you'll see in this example, knowledge of the radio plan and radio network will help the planners make better decisions.
Let’s say we have an area with 16 houses, the houses are distributed, and that the distances require the use of multi-hop in order to reach one of the two masters in the area.
In this very simplified low-scale example we'll show why the direction of the rollout is significant if you want the service personnel to verify the communication before moving on to the next house to install a new smart meter.
The radio planning tool takes into account distance, terrain, radio properties and many other parameters when suggesting how to optimize the network coverage, capacity and redundancy.
It's important to emphasize that the mesh network has powerful ways to find alternative communication routes. In this example we focus on the optimal communication route based on radio coverage calculations. This optimal communication route is not the only one, but is considered to be the strongest and most reliable. This is important when planning and when we want to make sure to get a strong network everywhere.
Analyzing alternative communication routes, redundancy, and simulating failure situations will be covered in later blog articles.
In this example the result of planning the optimal radio network in this area is logically like this:
In the schematic above we see that the slave nodes are placed from 1 to 4 hops from its optimal master node. The numbering is a notation used to easier describe the suggested communication route and the different roles of the different nodes.
The nodes in these two node trees are colored based on the number of hops from the master node.
- Green = 1 hop
- Orange = 2 hops
- Yellow = 3 hops
- Dark Cyan = 4 hops
The red node is identified as unreachable in this example, and will be planned in more detail at a later stage. There are a number of ways to extend the range or find alternative backhaul communication links for this type of outliers.
When looking at the same network and the geographical position of the same nodes we see that the distribution and distances are quite different from the logical topology above.
The gray circles are used to illustrate the groups of nodes based on the number of hops. The terrain and topography is not shown in this example, but was taken into account during the calculation of the optimal communication routes.
By adding road axis AB and CD to the theoretical map we see that the position of most houses map quite well to the roads.
For the planners there are many factors, local conditions and restrictions to consider when planning the rollout. In this example we only consider the radio-based factors and the road axis.
We assume that the master nodes are already built and that they are operational.
In this example it's quite obvious to see that all other orders of rollout than ABCD is less optimal assuming that this area is covered by one service van alone.
As a simple exercise use the map to consider the consequence of going DCBA.
In this example we have suggested a rollout sequence, denoted as a white number inside each of the nodes, that is close to optimal for this example when we only consider radio and road axis.
Just imagine the complexity of planning this at a large scale for many hundred thousand or millions of meters.
Being able to verify communication all the way back to the central system on site after installing the new meter is key in order to reduce the number of revisits.