Modifying Grover's Quantum Search Algorithm to Perform Effectively in the Presence of Noise

Since nearly the beginning of the 21st century, scientists have been tinkering with quantum computers that are able to output a solution based on data received. One issue that has arisen is the time it takes to search for words in an unsorted dataset, as the computer has to search one-by-one. Grover's Search Algorithm gives the solution of searching multiple possibilities at once, changing the time it takes to search N words from O(N) to (√N). However, the algorithm is very sensitive to noise when running on present-day quantum computers. To fix this problem, a trimmed algorithm was introduced, but that had a tradeoff of robustness for accuracy. Here we introduced a new algorithm that retains a similar time efficiency to Grover's Algorithm, while being able to improve its probability of success in completing the search. This was accomplished by replacing Grover's quantum diffuser with single (U) and double (RXX) rotational gates on every pair of qubits, along with a slight modification to the oracle to use multi-control Pauli-X gates rather than a control-NOT gate. The key results found from this research was that the introduced algorithm was able to outperform Grover's Algorithm for 3, 4, and 6 qubits, while keeping the time advantage.