TelecommunicationsBI, AI and ML for the business
Antenna retilt, before and after
Two drive-test campaigns on the same route, before and after the antennas were retilted. Muvia compares them lap by lap and stretch by stretch: SINR, received power and throughput tell you whether the work paid off, and on which stretches.
Illustrative scenario: it describes a typical case, not a customer project.
The antennas have been tilted. Did it help?
A retilt is decided on complaints and counters, carried out in one night and often judged on gut feeling. The drive-test measurements exist, but two campaigns two weeks apart end up in two files and two slide decks.
In Muvia the campaigns sit in the same source, and the comparison is made on the same stretches at the same hours. A distribution, the overlaid lap and a stretch-by-stretch table settle the question.
- Who it's for
- Radio optimisation engineers, drive-test teams, regional network managers and whoever signs off work on the antennas.
In Muvia, step by step
Real product screens, recorded on a project with sample data.
The video · Two drive-test campaigns on the same route, before and after the antennas were retilted. Muvia compares them lap by lap and stretch by stretch: SINR, received power and throughput tell you whether the work paid off, and on which stretches.
What you get
- A verdict, not an impression
- Distributions and a stretch-by-stretch table say whether the work improved the network, and where it did not.
- Like-for-like comparisons
- Same road, same hours: the differences come from the work, not from the route.
- All campaigns in one place
- Each new campaign joins the earlier ones and can be compared with all of them.
- The next job starts here
- The notebook stays in the project, with the figures and the decision taken.
For the technical team
How it is built in Muvia
- 1
Load the campaigns
The files of the two campaigns enter the same source, with a column that tells them apart. Each new campaign is appended to the earlier ones with the same columns.
- 2
Compare the distributions
In Analysis a condition separates campaign B from A: the SINR distribution, the day-by-day box plot and the XY view of RSRP against SINR show the shift.
- 3
Overlay the same lap
Comparing with an earlier period lays today's lap over the one 14 days before: same road, same hours.
- 4
Settle it stretch by stretch
A notebook tabulates the SINR and throughput gain on every stretch, with the conclusion, and prints to PDF for whoever approved the work.
- The data it needs
- The drive-test van's measurements, one sample every 5 seconds: stretch, serving cell, RSRP, SINR, throughput
- The campaign before and the one after the work, on the same route
- The cell and stretch inventory, from an Excel file
- The date and time of the work, as an annotation
- Parts of Muvia used
More cases in this line
All use cases- A cell's evenings of interferenceRadio cell counters arrive minute by minute: SNR, received power, users and dropped calls. A condition isolates the evenings when SNR collapses while power rises, Layers aligns them on their start and the heatmap shows the hour it happens.
- A suspect lot, contained in minutesMaterial lots, orders, serialised parts, sequenced shipments and the customer's vehicle numbers linked in one path. When a supplier flags a suspect lot, the exact scope to contain comes out in minutes, instead of a blind recall.
- Sequenced deliveries that never stop the customer's lineCar makers' call-offs, JIS and JIT shipments and delays in one dashboard, with OTIF per customer and a map of when late call-offs arrive. Logistics sees the risk of a line stop before it turns into a penalty.
Bring us a question you can't answer today.
We start from a real question your business has and walk the path from source to dashboard with your systems, not a demo dataset.