Posts

AI, RAN and Delay-Doppler - How Cohere does it

Image
Agreed. The title has too many nouns! Each of the nouns is important though, as we will find out. Allow me to explain. So this article is a profile of Cohere Technologies . Cohere features in our report “AI and RAN – How fast will they run”. As you would now guess, Cohere Technologies specializes in using the Delay-Doppler channel model. It uses this model in software to MU-MIMO enable existing FDD and TDD networks. What does the model do? This model leverages AI/ML for detection, estimation, prediction and precoding of channels. What does the term ‘Delay-Doppler' mean? It is easy to misinterpret this phrase as a special use-case of the Doppler effect – it is not. Cohere uses the word ‘delay’ as a shorthand for distance and the word ‘Doppler’ for velocity. Cohere postulates that tapping the channel simultaneously for time delay and Doppler effect provides a unique profile of the signal path from the antenna to the user – specifically, the scattering of signal and its causes can b

What Drives AI in Network Optimization Globally?

Image
Share of Addressable Market in Traffic Optimization End-Application Mobile RAN for AI and Related Technologies; by Geographical Region 2023-2028 Let us come straight to the point. RAN shipments data recorded globally by several sources reveals the CAGR to be in single digits. Insight Research, on the other hand, pegs the growth for addressable market for AI in RAN in healthy double-digits. What explains the apparent dichotomy? Well, the answer is very simple. There is no dichotomy here. Insight Research is not covering the market for the entire RAN, it is focused only on a high-growth niche driven by AI in the RAN. For AI applications covered by us, the base market size is low, the technology is new, the growth rate is bound to be higher. It is elementary really! Even then, one would like to know, where is this growth coming from? Where are the real world examples of AI being used in the RAN? Let us consider one major end-application of AI in RAN – Traffic optimization. For markers of

The ‘Marvellous’ Opening of the AI Accelerator in the 5G RAN

Image
  In end-February 2024, Marvell captured large bits of popular imagination when it open sourced OCTEON 10 ML/AI Accelerator Software to what it termed as optimization of 5G RAN Networks. Lying as it does at the intersection of AI and RAN, it is worth taking a deep dive into this development as AI in RAN remains an area of continued focus for Insight Research. It is worth noting that Marvell’s entry into the 5G domain was a result of the widely reported misfiring of Intel’s 10 nm promise to Nokia and Nokia’s subsequent unsuccessful tryst with Xilinx FPGA. That was 2019 and that was custom Silicon. This is of course, 2024; and we are taking about open-source accelerators. How did we get here? With a history of close to three decades, Marvell’s tryst with telecommunications is not new. Apart from in-house development, Marvell has over the years acquired companies engaged in ethernet switches, embedded networking software, 3G SoCs, IMS software, network processors and data center switc

The AI-RAN Alliance: THE idea whose time has come

Image
  Of all the announcements at MWC24, the one that validates our focus on AI in RAN is the formation of the AI-RAN Alliance . Coinciding with the release of our report “AI and RAN – How fast can they run?” , the Alliance checks all the boxes to furthering AI from the periphery of the RAN to its very core (pardon the pun, not to be confused with the mobile core). The Alliance has trifurcated the engagement of AI with the RAN at the following levels in its demonstration at MWC24 :   AI on RAN deals with making the RAN work for furthering AI end-applications. It is well understood that the availability of high-quality and high-throughput bandwidth will help improve the reach and spread of AI. This approach has a low barrier for entry as the emphasis is primarily on harnessing existing RAN infrastructure. Here, AI is the tenant, and RAN is the provider.   AI for RAN turns the earlier tenant-provider equation upside down and looks at AI as a tool for improvement in the RAN functioning. T

AI and its applications in the RAN

Image
  Excerpted from our report AI and RAN – How fast will they run? The above figure charts the progression of the revenue shares of the key end-applications for AI in the RAN. Insight Research identifies the following key end-applications for AI in the RAN ·       Traffic optimization ·       Caching ·       Energy management ·       Coding The impression surrounding the all-pervasiveness of AI is principally shaped by the somewhat recent unleashing of the power of generative AI. Generative AI and LLM are arguably the more flamboyant exponents of AI and definitely the more recent ones. AI as a construct, however, has its roots in the previous century. AI has made itself at home in the RAN almost as early as the advent of cellular mobility itself. There have been several compelling reasons for the technology to operate under the radar for a major portion of its existence: There was little appreciation of AI as a distinctive technology. This was because AI offered only piecemeal increment

Decoding the RAN coding conundrum - with AI!

Image
  How important is coding in the RAN? The answer, “very important” is not going to win you any awards, as it is a no-brainer! Coding is not just about data representation; it's about optimizing data flow, enhancing error correction, and ensuring robustness in diverse network conditions. Let us look at a very specific intent – error correction. To prevent data from being corrupted by channel interference and noise, channel coding employs redundancy. At the heart of 5G RAN's coding mechanisms is the use of Polar Codes and LDPC (Low-Density Parity-Check) codes. Polar Codes work well in 5G control channels. This is because, they are well suited for limited or finite data lengths.   LDPC codes works well with longer lengths. Not surprising, given the “low density” properties of the codes. LDPC essentially ensures that even if some packets are corrupted during transmission, the original data can be reconstructed without retransmission. LDPC codes, are therefore employed for d

“En-caching” the RAN – the AI way

Image
  RAN caching is an intuitive use-case for AI. Our report “AI and RAN – How fast will they run?”, places caching third in the list of top AI applications in the RAN. There is seriously nothing new about caching. In computing analogy, caching is as old as computing itself. The reason caching and RAN are being uttered in the same breath is primarily MEC. MEC is a practical concept. MEC attempts to leverage the distributed nature of the RAN infrastructure in response to the explosion in mobile data generation and consumption. Caching looks at practically every point in the RAN as a possible caching destination – it can be base stations, or RRHs, or BBUs, or femtocells, or macrocells and even user equipment. The caching dilemma is a multipronged – what to cache, where to cache and how much to cache. In an ideal world, one could have access to infinite storage and processing capacity interconnected with infinite throughput at zero latency. In the real world however, each of these a