Most of us rely on classical computers to do analysis, perform calculations, and store information every day. Scientists and engineers write programs to solve difficult problems and run complicated simulations. But after decades of development, it is getting harder to make smaller transistors to build faster and more energy efficient classical computers. For supercomputers, though they are very powerful machines, there are still intractable problems, which means no efficient algorithm to solve them.
The field of quantum computing was initiated in 1980s and became practical after 2006 due to faster classical processors and more capable Field Programmable Gate Arrays (FPGA). Like classical computers, quantum computers have quantum bits (qubit) to carry information. The single classical bit can only be ‘1’ or ‘0’, while quantum bit can be BOTH ‘1’ and ‘0’, which is one important quantum mechanical phenomenon called superposition. For example, in our current 64-bit PC, one 64-bit register stores 1 value. In quantum computer, the 64-qubit can represent 2^64 values at the same time, and it is the quantum superposition effect at the atomic level.
The other important quantum mechanical phenomenon is called entanglement, which means the qubits are no longer independent and there are interactions and behavioral links between them. These two phenomena help to build a set of linked qubits, which evolve inside the quantum processor with the predefined algorithm and the given input state.
A simplified semiconductor quantum computing physical layer is illustrated in Figure 1. The high-speed D/A generates the microwave signal (4–10GHz) to initiate the initial state of the qubits. The qubits evolve to all possible results states, all existing at the same time inside the processor. At the instant we measure
the qubits using high-speed A/D, all possibilities collapse into a single result to be read out from the processor. Then we can perform “initiate and measure” for multiple times to find out the most probable answers for a difficult problem.
Innovative Integration has been a leading supplier in the real-time embedded system market for more than 30 years, and we are proud to be the control and measurement system provider to the key players in the quantum computing field. As shown in Figure 2, using the high-speed, low-latency X6-1000M modules inside the compact VPX PC, we can configure the synchronous system to control and measure up-to 32 entangled qubits. With the state-of-the-art FPGA technology, engineers can perform FFT and generate waveforms to sync with other qubits within 250 ns. This low latency system helps to perform as many operations as possible before quantum decoherence occurs.
• 16-qubit compact VPX control and measure system; potentially up-to 32-qubit
• Synchronous, high-speed, low latency 32x 1GHz, 16-bit D/As; and 32x 1GHz, 12-bit A/Ds
• Low latency crossbar switch for qubit synchronization; 4 digital IOs per qubit
• Built-in precise digital sequencer for peripheral devices control
• One Xilinx FPGA per qubit for real-time DSP/FFT/state estimation
• Graphical FPGA firmware devkit using Matlab/Simulink and Xilinx System Generator
• NOT include external microwave sources, mixers, filters, attenuators, refrigerator
It is challenging to build a quantum computer to achieve quantum supremacy, which means to solve the problems that the classical computers cannot. It requires at least 49 high quality and entangled qubits. Both Google and IBM are claiming that they are building and testing 50 qubits by the end of 2017. We believe the first large scale and general-purpose quantum computer will hit the market soon. It will greatly change our way to do encryption, database searching, new medicine and materials, especially when combined with artifact intelligence (AI).