Optimization – The Missing O in Open RAN

This Industry Viewpoint was authored by Eric Dowek, Segment Marketing Director at AccelerComm

The Open RAN movement has generated considerable interest from across the 5G market, promising to bring significant benefits to the mobile operator and vendor ecosystems. The opportunity is significant, with analysts such as Omdia predicting Open RAN will be a $3.2B market in 2024.

Open RAN draws on the same ethos that made the mobile industry a global success in the 2G era – open and collaborative co-operation on standards that will ensure interoperability of equipment. GSM’s success was based on collaboration between the entire value chain to the benefit of the whole world.

Open RAN has a similar potential to catalyse 5G innovation. It holds the power to create a rich and broad ecosystem of vendors, with specialists in various components collaborating to drive technical breakthroughs and developments. Importantly, Open RAN also offers the promise of reducing the possibility of operators being locked-in to working with a single vendor and will help to create more competition in the infrastructure market. Yet, one component seems to be missing in Open RAN’s success formula: Optimization.

Open RAN comes with its set of challenges

Whilst much of the work in the Open RAN community is, rightly, predominantly focused on interoperability between vendors, optimizing the performance of solutions is also critical and should not be overlooked. For Open RAN networks to be competitive they must offer the same, or better, performance as end-to-end solutions from a single vendor. If the radio access network performs below par, they will require more infrastructure to deliver high quality services, increasing CAPEX and OPEX, or they will be forced to offer lower value services to end-users. Either of these outcomes would mean decreased ROI for the operator and ultimately stifled growth of Open RAN and 5G itself.

Optimizing Open RAN networks

To fully realize the benefits of Open RAN, any solution needs to be optimized. As an example, let’s take the radio network itself: The spectral efficiency of a radio network (bits/sec/Hz) is a key performance indicator that defines how efficiently an operator is using the spectrum that in which they have invested. The receive sensitivity of the gNodeB can limit the uplink, which in turn directly impacts the spectral efficiency for both the uplink and downlink. Recent research from the industry analyst firm Mobile Experts quantified the impact on a typical US mobile operator. They calculated that the impact of halving the gNodeB receive sensitivity would be to increase the number of gNodeBs required in a typical Open RAN US network by 2550, adding approximiately $193m in infrastructure costs. 

In this example, the cause of the low receive sensitivity could be the use of generic channel coding/decoding solutions. At the core of the 5G physical layer is a Low Density Parity Coding (LDPC) decoder. LDPC is a linear error correction scheme which helps to clean things up such as noise and interference that get in the way of reliable 5G data transmission. The algorithm and architecture used to implement an LDPC decoder can change its performance dramatically. Put simply, relying on generic products, often derived from Wi-Fi or broadcast products, significantly impacts the performance of the network. The result is reduced spectral efficiency; the impact is more equipment needed to provide the same network performance. 

In contrast, a complete 5G channel coder/decoder design that includes a highly optimized 5G LDPC core, together with other critical components such as the channel equalizer and estimator, can bring dramatic benefits, doubling the receive sensitivity compared to generic solutions.

To unlock the full benefits of Open RAN, the focus must now be on the missing component of its formula: optimization. The channel coding example is just one component in the RAN value chain, but it is an excellent example of the pitfalls in not focusing on optimizing these networks to realize their true potential.

In order to succeed, Open RAN will require vendors to concentrate firmly on both optimization and interoperability.  If they can achieve this, and bring together best in class vendors from across the ecosystem to deliver Open RAN networks that outperform the alternatives, the upsides are significant.

If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!

Categories: Wireless