Blog ENG

Enhancing Network Coverage Planning

Josipa Jurić

A top-tier telecom and SQream POC: Enhancing network coverage planning efficiency

A Top-tier telecom faced the challenge of their mobile network coverage calculation and other relevant layers for network planning because they require a lot of manual work, 3rd party projects, etc. These time-consuming processes were slowing down the implementation of IT projects throughout the organization.

Their solution was developed in Python with constant ad-hoc manual work. The network planning department had limited manual resources to service these requests.

A Top-tier 1 telecom company needed robust, scalable, and innovative technology for the BAU and new sustainable solutions. Their target audience was Master Plan Department (Network).

A Top-tier 1 telecom company collaborated with SQream to enhance their network coverage calculations. They both presented Post POC presentation of the HyNetMatrix solution to the IT Accelerator.

About the HyNetMatrix solution

The main goal of the HyNetMatrix solution based on SQream technology was to significantly reduce time-to-market and improve the accuracy of planning.

The diagram below illustrates how a Top-tier telecom and SQream’s solution works.

The plan was to increase performance from 2 days to ~2 hours per query with SQream, to ingest massive data (3TBH/GPU) and deliver near real-time analytics.

It was planned to achieve that goal both on Premise with SQream on a Cluster of GPUs based on the x86 platform & IBM Power9 platform and on Cloud (GCP) with SQream.

POC Scope:

  • Ingest 800 MHz and 2.1 GHz bands from LOS files
  • Generate RSRP raster and Cell Polygon for the 800 MHz and 2.1
  • GHz bands with resolutions of 100 meters for the 800 MHz band and 25 meters for the 2.1 GHz band
  • Generate RSRP raster in 100-meter resolution for both 800 MHz and 2.1 GHz bands
  • Incremental updates
  • On-Prem & GCP (Hybrid)

Results

The project results were impressive. It achieved ingest of 10.5 TB in 50 mins with ingestion speed 3 TB/h/GPU, 4 GPU in use. The average Compression Ratio is 1:8 (storage).

The results of the proof of concept (POC) were outstanding, as evidenced by the table below.

Stakeholder validation

Frank Meisgen, FBE, one of the stakeholders of the IT Accelerator, concluded after the presentation: ,,Impressive how the massive parallel computation power of Google Cloud can still be improved by using GPU to tackle network optimization problems.”

Why SQream?

SQream is a SQL database that empowers organizations to perform complex analytics on a petabyte-scale of data and gain time-sensitive business insights faster and cheaper than from any other solution.

Megatrend poslovna rješenja are SQream partner on European market. Find out why organizations ranging from fast-growth startups to Fortune 100s all rely on SQream.

For all your additional questions, contact our team for a presentation meeting at +385 1 4091 200 or poslovna.rjesenja@megatrend.com.