Network Drive Testing
DRIVE TEST TYPES
Depending on the motivation for testing, there are variations in the types of network drive tests that can be performed:
- Benchmark testing (side-by-side, multi-channel performance evaluation of two or more co-located networks)
- Troubleshooting (identifying the root-cause of network malfunctions or performance degradation)
- Optimization (improving network performance through manipulation of network parameters)
- Conformance (verification of network adherence to regulatory requirements or minimum performance standards)
DRIVE TEST SERVICES
At GHB Intellect, we provide the expertise, the resources, the tools, and the turn-key management to perform all types of network drive tests and at every stage, from planning and data collection to data analysis and optimization recommendations. Our strengths include:
- Expertise in air interface standards
- Expertise in network operation and optimization
- Experience with various drive test equipment and post-processing tools
- Expertise in drive test procedures and best practices
- Expertise in data processing and analysis
These all work hand-in-hand to cost-effectively provide the best service quality to our clients. Below is a brief description of what services we provide related to network drive testing.
NETWORK DRIVE TEST PLANNING
The most crucial part of an overall network drive testing project is the test planning stage. A proper test plan provides a detailed procedure for:
- What drive test and data processing equipment/tools to acquire that cost-effectively collects and analyzes the necessary data and produces the sought-after reports
- What PHY, MAC, or Network level signals and parameters to collect
- What procedure to follow in collecting the data to induce the conditions that require monitoring
- What drive routes, resources and timelines to consider for performing the drive tests
- What are the logistics related to collecting and storing the data in safe and secure manner
- How to efficiently process the collected data
- How to best analyze the collected drive test data and draw the necessary conclusions
The proper performance of drive testing, based on the test plans developed in the previous step, is also important. As the old adage goes, “garbage in, garbage out.” If meaningful data is not properly collected, it is possible to draw incorrect conclusions from them. Incorrect or incomplete data collection can also lead to additional drive testing effort, increasing the overall cost of the project. As such, it is important to adhere to proper data collection procedures and best practices. This is where experience in field work and network drive testing can be quite instrumental. Also, important is experience with handling various drive testing equipment.
Even though we have so far been focusing “drive” testing, the network performance evaluation and comparisons can nowadays take many other forms. This includes indoor, pedestrian, train, underground, marine and other types of network measurements.
DATA POST-PROCESSING AND ANALYSIS
It is one thing to perform a set of network drive tests and collect all the required data points tagged with time and GPS location, and it is another thing altogether to sift through the myriad of data and hone in on the data of interest. Depending on whether what type of data was collected the size of the collected data may vary tremendously. For example, the size of data captured at the Physical Layer (PHY), such as I and Q signals, could easily grow extremely large. Without the proper plan and tools to manage and examine such a vast collection, the processing task can be quite cumbersome, time-inefficient, and inconclusive. The drive test plan of step 1 comes very handy here. However, experience with handling such data is also important.
Once the interesting subsets of the collected data are identified, analysis of the data becomes the next main task. The analysis could be straight forward or quite complicated. For example, if the goal of the drive test is to verify certain functionalities in the network are present, it may be sufficient to simply review the values of certain network parameters (usually layer 3 parameters) along a drive route. On the other hand, there are times that the performance of the network under certain conditions is to be analyzed, or the correlation between two or more different functions in the network is of interest. In such cases, no direct relationships may be available and statistical analyses may be the only option to pursue. Having the experience and the proper statistical model to approach the problem becomes extremely important.
Finally, once the data is analyzed, presentation of the analyzed data in a manner that is easy to follow is not a trivial task. The report generation must be done so as to allow the reader to follow the entire process of planning, drive testing, and analysis. Moreover, the pertinent data needs to be presented in various forms (geographical, tabular, claim charts, etc.) to support the conclusions being made.
EQUIPMENT AND TOOLS
At GHB Intellect, we have years of experience in utlitizing a wide array of equipment and tools for network drive testing, data collection, and data analysis. Some of the tools and equipment we have used in the past are listed below.
GHB Intellect is independent of equipment/tool manufacturers and network operators. As such, our clients have utmost confidence that they receive completely unbiased and proficient consultation and services.