The recent performance report published by the Telecom Regulatory Authority of India (TRAI) indicates that the contribution of 4G and 5G data usage in total volume of wireless data usage are about 79.17 and 20.26 per cent respectively, indicating the huge uptake of these services.

However, the quality of service of even plain old telephony on our mobile networks continues to be very poor. We continue to get call drops, especially inside buildings and on highways. According to the recent Ookla Speed Test report, India slipped to 26th position in the average mobile broadband speeds with 92 Mbps and 8 Mbps as the downlink and the uplink speeds respectively.

The TRAI initiated Quality of Service (QoS) regulation in 2000 and revisited the measurement and metrics in 2009, and 2012. However, despite these measures, the overall quality of experience often is not good enough for consumers. Realising the same, TRAI floated a consultation paper on developing new norms and metrics for mobile broadband service in 2023.

Root causes

The moot question is what is the root cause for the poor network quality?

The signal quality is often very poor in rural and remote areas due to the cellular tower in which the Base Transceiver Station (BTS — also called as eNodeB in case of 4G networks) is often far from the mobile handset. The signal quality drops as the distance increases.

Further, depending on the spectrum that is used, the propagation of radio waves inside buildings can also be very poor leading to poor signal quality, which in turn affects voice and data transmission. The high frequency radio waves often have poor propagation characteristic compared to low frequency spectrum (often referred to as sub GHz spectrum).

The Mobile Network Operators (MNOs) often use the two sides of the coin: spectrum and the BTSs to provide coverage and quality of service. If the spectrum held by the operator is less or that the MNO does not have adequate low frequency spectrum, then they should deploy more cell sites and BTSs to cover a specific area.

However, the cellular towers and the associated radio equipment are capital intensive. Most of the MNOs often lease this access infrastructure from tower companies such as Indus Towers, which results in higher leasing and rental costs.

Trade offs

The MNOs do a trade-off between acquiring spectrum in the appropriate band versus the deploying cell sites and BTS. This typically results in higher prices for the low frequency bands such as 700 MHz and lower prices for high frequency band such as 3.3 GHz and 26 GHz bands. The average winning bid prices of 700 MHz, 3.3 GHz and 26 GHz in the 2022 spectrum auctions were $0.32, $0.054, $0.0012 per MHz per capita. In Delhi, the reserve price for 700 MHz price was $1.84 per MHz per capita, much higher than in many countries.

One of the main factors determining the winning bid prices in spectrum auctions is the reserve price or the starting bid price at the beginning of the auction. In India about 2 X 35 MHz per Licensed Service Area and hence a total of 1,540 MHz was put on block in the spectrum auction held in October 2016.

However, due to no bidding activity in this band, possibly due to high reserve prices at an average of $1.16 per MHz per capita, the whole lot went unsold. The subsequent auctions held in 2021 also witnessed no activity in this band. Finally, in the auction held in July 2022, about 2 X 10 MHz per LSA was picked up by one of the MNOs across all LSAs at an average winning bid price of $0.32 which is equivalent to the reserve price in most of the LSAs.

However, the metro areas of Delhi, Mumbai and Kolkata witnessed a very high price of $1.84, $1.54, and $0.74, respectively. In effect, a large amount of spectrum in this band has been unused for the last eight years! Further the 600 MHz band of which 1,320 MHz across the country was put into auction in 2022 remained unsold.

Hence, it is time to re-look at the reserve prices of these sub GHz bands in the future auctions. Research indicates that the number of cell sites required to provide equivalent geographic 5G speed coverage with 700 MHz is about one-fourth with that using the mid-band spectrum such as 1800 MHz.

The availability of 600 and 700 MHz at appropriate price points will encourage the MNOs to acquire enough spectrum for improving 5G penetration especially in rural areas. Even in urban areas, the low band spectrum can provide superior in-door coverage thereby augmenting the widely used mid band (3.3 GHz band).

Formula tweak

It is time that we depart from the formula of fixing reserve prices based on the winning bid prices in the previous auction. This method provides no room for decreasing the reserve price. We can use Simultaneous Multiple Round Ascending (SMRA) auction, invented by the Nobel Laureates Milgrom and Wilson, with enough guard rails that is proven to induce “truthful bidding”.

If we reduce the reserve prices to allow the methodology and the market to determine winning prices, we are sure to witness the MNOs picking up the yet unsold 600 and 700 MHz spectrum bands. Only then that we will be able to realise the true benefits of 5G and beyond networks.

Sridhar is Professor; and Mahaadevan is an M.Sc student at IIIT-Bangalore

Related Topics