Reliance Jio data speed not as fast? Debate continues over TRAI biasness

Adding a new twist to its stance of alleged bias of telecom regulator TRAI towards Reliance Jio, cellular body COAI (Cellular Operators Association of India) has raised questions over the data captured by TRAI's MySpeed app. In a letter to the TRAI secretary, the COAI has asked TRAI to review its February 5 white paper on Measurement of Wireless Data Speed.

Since the rollout of 4G network, the data speed has become the talking point. The telecom operators are thumping their chests on offering the highest 4G speeds quoting different speed test results.

There are four speed tests at the moment - TRAI's MySpeed app, Ookla, OpenSignal and 4G Mark. Test results for these apps are crowd-sourced from regular users under conditions of normal usage. In addition to sourcing data from users, TRAI also conducts independent drive-tests (IDT) across various parts of the country to determine the data speeds of various network providers.

COAI has raised objections on the discrepancies with the TRAI's MySpeed app and its IDT results, the locations at which speed tests are conducted and the sample collection process on the MySpeed app. For instance, the COAI found that test samples collected for the app were located on water bodies for one of the telecom operator (read Jio). Those particular samples were found to have speeds above 10 megabits per second (Mbps) which COAI says is practically impossible.

"Certain discrepancies within the results of the TRAI MySpeed app primarily question the level of accuracy of this data and also may defeat the purpose of creating of such an app," COAI noted in its letter to TRAI.

In another instance, COAI says that variations among the published results of MySpeed app and IDTs is visible for several towns and in certain cases, the variation in MySpeed app is very high in case of one telecom operator (read Jio) as compared to IDTs.

Strangely, agencies like OpenSignal stay away from collecting data from test drives because of its limiting factor. The test drives rely on employing a limited set of devices in a small number of locations which cannot capture the capabilities of an entire network. On the other hand, crowd-sourcing relies on measurements from millions of smartphones owned by normal people.

The issue seems serious because the possibility of "fixing" the test results - assuming that is the case - in case of IDTs is far easier than app. It's possible to rig a few samples than manipulating data collected from millions of users spread across the country. Until TRAI responds to COAI's letter (if it chooses to), the curious case of 4G speed tests will remain unsolved.