TOKYO—Toshiba Corporation, a pathfinder in quantum technology, has applied it quantum-inspired optimization solution, the Simulated Bifurcation Machine (SBM) (*1-4) to optimal time and frequency allocation for 5G communications, and, for the first time ever, succeeded in allocating resources to 20 terminals in less than 0.5 ms. This achievement surpasses the minimum transmission delay of 1.0 ms expected in the 5G standard (*5).
5G delivers high speed communications, low latency, and multiple simultaneous connections. However it has proved difficult to allocate resources for simultaneous, low latency communications by multiple terminals within the standard. Toshiba’s resource control technology realizes an efficient communications environment that can simultaneously secure low latency communications with multiple terminals, which will contribute to the application of remote control and automation of equipment at industrial sites such as factories and warehouses (Figure 1). Toshiba expects the algorithm will be used in cell phone base stations and various other sites that deploy 5G, and will market widely to those businesses as a way to contribute to the promotion of digital transformation.
Toshiba plans to present the technology at IEEE WCNC 2024 (*6), an international conference in Dubai, United Arab Emirates, from April 21 to 24, 2024.
Development background
Many countries are seeing declines in their working-age populations, and manpower shortages are making it increasingly difficult for factories and warehouses to pass on knowledge and skills. One widely adopted approach to solving this problem is to apply IoT to the automation of cargo transportation, and the use of robots to improve business processes. The efficient use of multiple robots without bottlenecks and other issues requires simultaneous, real-time communications and control. Local 5G, the operation of 5G networks by non-carriers within a limited area, such as a building or piece of land, has emerged as an important solution, one that is expected to spread in the future.
Achieving the high-speed, low latency, and multiple simultaneous connections necessary for instantaneous decisions on which resources to allocate to multiple terminals must meet numerous requirements, including quality of communications. However, it is difficult to achieve flexible resource allocation while achieving the expected minimum transmission delay of 1.0 ms. Most notably, it has proved extremely difficult to handle the requirements of multiple terminals and carry out exhaustive searches for optimal solutions for resource allocation simultaneously and in a timely manner.
Features of the technology
Toshiba has achieved low latency in resource allocation with SBM. The algorithm solves combinatorial optimization problems and finds optimal solutions from huge numbers of alternatives at ultra-high speed, and has already demonstrated its practical value in fields as diverse as finance, drug discovery, AI, and wireless communications. The resource control algorithm developed by the company performs two-step optimization: once SBM has solved a resource allocation optimization problem under conditions that relax the resource allocation constraints imposed by the 5G standard, the solution and statistical information are then used to further optimize the resource allocation within those constraints (Figure 2).
In an evaluation using the SBM algorithm, Toshiba randomly placed 20 terminals on a simulator, calculated their channels, and fed the results to SBM to find a solution. The process was repeated 1000 times and used to verify the resource control algorithm applied to SBM. The results confirmed that SBM provides 100,000 times better quality allocation than that realized by applying Greedy (*7), the currently used allocation algorithm (Figure 3). This is because SBM performs global optimization, but Greedy is capable only of local optimization. SBM also achieved the major milestone of the world’s first confirmation of allocation within 0.5 ms for all sample problems.
This development is expected to improve workplace efficiency. For example, it enables robots introduced to automate industrial settings to operate at speeds as fast as a few milliseconds.
Future developments
High quality 5G is expected to find use in industry, in areas such as high speed, precise control of multiple robots. As SBM can optimize huge numbers of parameters at ultra-high speed, it is expected to find application in resource allocation and wireless communications. By widely utilizing SBM, Toshiba will realize 5G that delivers high speed, low latency, and multiple simultaneous connections, and contribute to solving social issues such as productivity improvement and service enhancement, including the elimination of labor shortages through the promotion of DX.
Figure 3: Better allocation than conventional allocation explored in 100,000 iterations
*1: https://www.global.toshiba/ww/technology/corporate/rdc/rd/topics/19/1904-01.html
H. Goto et al., “Combinatorial optimization by simulating adiabatic bifurcations in nonlinear Hamiltonian systems,” Science Advances 5, eaav2372, 2019.
https://doi.org/10.1126/sciadv.aav2372
*2: https://www.global.toshiba/ww/technology/corporate/rdc/rd/topics/21/2102-02.html
H. Goto et al., “High-performance combinatorial optimization based on classical mechanics,” Science Advances 7, eabe7953, 2021.
https://doi.org//10.1126/sciadv.abe7953
*3: https://www.global.toshiba/ww/technology/corporate/rdc/rd/topics/21/2103-01.html
K. Tatsumura et al., “Scaling out Ising machines using a multi-chip architecture for simulated bifurcation,” Nature Electronics 4, pp. 208-217, 2021.
https://doi.org/10.1038/s41928-021-00546-4
*4: Quantum-Inspired Optimization Solutions, SQBM+™, provided by Toshiba Digital Solutions Corporation
https://www.global.toshiba/ww/products-solutions/ai-iot/sbm.html
*5: World's first optimization of resource allocation for multiple terminals in less than 0.5 ms, which is required to achieve the minimum transmission delay of 5G specifications (Toshiba research)
*6: H. Obataet al., “Ultra-High-Speed Optimization for 5G Wireless Resource Allocation by Simulated Bifurcation Machine,” IEEE Wireless Communications and Networking Conference (WCNC), Apr. 23, 2024, Dubai,
https://wcnc2024.ieee-wcnc.org/
*7: In telecommunications, an algorithm that determines the order of terminals selected, starting with the channel with the best quality and continuing to select on that basis.