The advent of 5G technologies brings us to a new era in the wireless communication technologies field. We are witnessing a breakthrough that supplies us with faster speeds, lower latency, and higher capacity while having a higher energy efficiency than previous wireless communication technologies. It also unleashes possibilities for new applications in many fields, such as the Internet of Things (IoT), smart cities, logistics, and unmanned autonomous vehicles (UAVs).
This technological leap is only possible thanks to the intertwined contribution of many different technologies, each disruptive in its own right. Among those, Multi-User Multiple-Input Multiple Output (MU-MIMO) technology is an essential technique for wireless signal transmission for 5G technologies. This allows for an increase in transmission speed and the number of users that can be connected simultaneously. However, this technique presents some technical challenges, such as signal decoding, that require state-of-the-art hardware and algorithms to be tackled effectively and efficiently.
MIMO technology uses multiple antennas to send and receive data to improve channel capacity, allowing for higher data rates.
This technique requires decoding the received signal, which is very sensitive to channel noise. To overcome it, 5G communication systems typically have many antennas, but only a few users can use them simultaneously. By these means, increasing the number of antennas per user allows for averaging the noise and, hence, for easier recovery of the original signal.
To increase the number of users an antenna can support, we could rely on brand-new, enhanced MIMO detection algorithms rather than increasing the number of receivers. Thanks to these, 5G networks can deliver a better user experience by better managing complicated channel conditions and interference scenarios.
Moreover, this decoding procedure must be done quickly (typically less than 5 ms) to serve as many users as possible while having the lowest possible latency. Thus, decoding algorithms must be fast and inexpensive.
Currently, algorithms in use have very low complexity but are susceptible to channel noise in exchange. We solve that problem using a Stochastic Optimization approach in tandem with our brand-new Quside’s Randomness Processing Unit (RPU). Using stochastic algorithms may be a perfect trade-off between complexity and performance.
Stochastic algorithms can be very effective at finding solutions to problems that are hard to solve with traditional deterministic algorithms. Therefore, applying this stochastic algorithm to the problem improves the results of using a conventional deterministic algorithm.
Quside RPU is a perfect tool for stochastic optimization problems such as MIMO detection. Quside’s RPU shows faster simulation speeds and better convergence of their stochastic simulations, even compared with state-of-the-art algorithms.
The following picture compares a Stochastic algorithm and a conventional decoding algorithm, such as Zero-Forcing (ZF), commonly used in production-ready MIMO antennas. The left-hand side picture shows the Bit Error Rate using both algorithms for a MIMO system with 64 receiving antennas used by 16 or 32 users. Using the RPU reduces the BER in both situations independently of the channel’s noise. The right hand side picture shows the BER improvement achieved using the RPU with respect to only using the ZF algorithm. All the algorithms take less than 5 ms to perform the decoding.
With Quside’s RPUs and cutting-edge stochastic methods, we’re not just keeping latency low but also boosting transmission reliability, even in the harshest channel conditions. These improvements can imply multiple benefits to MIMO technologies, such as increasing the number of connected users or the transmission speed.
By leveraging the power of stochastic methods for MIMO detection, 5G technologies can enjoy some serious perks, such as boosted performance, reduced costs, unparalleled flexibility, and a significant competitive advantage.