#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ Delayed-Choice Quantum Eraser (Qiskit, AerSimulator) Author: ChatGPT Usage examples: python quantum_eraser_qiskit.py --shots 6000 --nphi 25 --plot --outdir results python quantum_eraser_qiskit.py --shots 4000 --nphi 13 Requires: pip install "qiskit==1.2.*" "qiskit-aer==0.15.*" matplotlib """ import argparse import math import os import csv from typing import Dict, List, Tuple import numpy as np from qiskit import QuantumCircuit, transpile from qiskit_aer import AerSimulator # -------------------------- Core circuit -------------------------- def eraser_counts_qiskit(phi: float, erase: bool = True, shots: int = 8000, seed: int = 1234) -> Tuple[Dict[str, int], List[str]]: """ Build and run the quantum eraser circuit for a given phase phi. Qubits: q0 = S (signal), q1 = I (idler); Classical: c0 = S, c1 = I Returns (counts, memory). Memory list contains bitstrings 'i s' (c1 c0). """ qc = QuantumCircuit(2, 2) # First beamsplitter on S qc.h(0) # Phase accumulated by S (analog "path phase") qc.rz(phi, 0) # Which-path marking via entanglement qc.cx(0, 1) # Second beamsplitter on S qc.h(0) # Choose measurement basis for Idler if erase: qc.h(1) # measure I in X (eraser); else Z (which-path) # Measure qc.measure(0, 0) # S -> c0 qc.measure(1, 1) # I -> c1 backend = AerSimulator(seed_simulator=seed) tqc = transpile(qc, backend=backend, optimization_level=3) job = backend.run(tqc, shots=shots, memory=True) res = job.result() counts = res.get_counts() memory = res.get_memory() # list of strings 'i s' (c1 c0) return counts, memory # ---------------------- Post-processing utils --------------------- def conditional_prob(counts: Dict[str, int], i_bit: str = '0') -> float: """ Compute P(s=0 | i=i_bit) from joint counts. Bitstring format is 'i s' (c1 c0), e.g., '10' means i=1, s=0. """ tot = n0 = 0 for k, v in counts.items(): i, s = k[0], k[1] if i == i_bit: tot += v if s == '0': n0 += v return (n0 / tot) if tot else 0.0 def hist_S_uncond(memory: List[str]) -> Dict[str, int]: """Histogram for S (unconditioned).""" from collections import Counter s_bits = [m[1] for m in memory] return dict(Counter(s_bits)) def hist_S_cond(memory: List[str], i_bit: str = '0') -> Dict[str, int]: """Histogram for S conditioned on I=i_bit.""" from collections import Counter s_bits = [m[1] for m in memory if m[0] == i_bit] return dict(Counter(s_bits)) def visibility(pvals: List[float]) -> float: """Compute interference visibility V = (max - min)/(max + min).""" if not pvals: return 0.0 pmax, pmin = max(pvals), min(pvals) denom = pmax + pmin return (pmax - pmin) / denom if denom else 0.0 # ----------------------------- I/O ------------------------------- def save_csv_prob(phi_list: List[float], p_i0: List[float], p_i1: List[float], path: str) -> None: os.makedirs(os.path.dirname(path), exist_ok=True) with open(path, 'w', newline='') as f: w = csv.writer(f) w.writerow(['phi', 'P(s=0|i=0)', 'P(s=0|i=1)']) for phi, a, b in zip(phi_list, p_i0, p_i1): w.writerow([phi, a, b]) def save_memory(memory: List[str], path: str) -> None: os.makedirs(os.path.dirname(path), exist_ok=True) with open(path, 'w', newline='') as f: for m in memory: f.write(m + '\n') def plot_curves(phi_list: List[float], p_i0: List[float], p_i1: List[float], title: str, outpath: str) -> None: import matplotlib.pyplot as plt plt.figure() plt.plot(phi_list, p_i0, marker='o', label="P(s=0 | i=0) (~ '+_X')") plt.plot(phi_list, p_i1, marker='s', label="P(s=0 | i=1) (~ '-_X')") plt.xlabel('φ (radiani)') plt.ylabel('Probabilitate') plt.title(title) plt.legend() plt.grid(True, linestyle='--', alpha=0.4) plt.tight_layout() os.makedirs(os.path.dirname(outpath), exist_ok=True) plt.savefig(outpath, dpi=160) plt.close() # ---------------------------- Runner ----------------------------- def run_scan(nphi: int, shots: int, outdir: str, seed: int, save_mem: bool, do_plot: bool) -> None: # Scan φ în [0, 2π] phi_list = list(np.linspace(0.0, 2.0 * math.pi, nphi)) # ----- Eraser mode (I in X-basis) ----- p_plus, p_minus = [], [] for idx, phi in enumerate(phi_list): counts, memory = eraser_counts_qiskit(phi, erase=True, shots=shots, seed=seed) p_i0 = conditional_prob(counts, '0') # i=0 ≈ '+_X' p_i1 = conditional_prob(counts, '1') # i=1 ≈ '-_X' p_plus.append(p_i0); p_minus.append(p_i1) print(f"[ERASE] phi={phi:.3f} P(s=0|i=+)={p_i0:.3f} P(s=0|i=-)={p_i1:.3f}") if save_mem: save_memory(memory, os.path.join(outdir, f"memory_erase_phi_{idx:02d}.txt")) save_csv_prob(phi_list, p_plus, p_minus, os.path.join(outdir, "eraser_probs.csv")) print(f"Vizibilitate ERASE V(+)≈{visibility(p_plus):.3f} V(-)≈{visibility(p_minus):.3f}") if do_plot: plot_curves(phi_list, p_plus, p_minus, "Quantum eraser (I in X)", os.path.join(outdir, "eraser_plot.png")) # ----- Which-path mode (I in Z-basis) ----- p_i0_list, p_i1_list = [], [] for idx, phi in enumerate(phi_list): counts, memory = eraser_counts_qiskit(phi, erase=False, shots=shots, seed=seed) p_i0 = conditional_prob(counts, '0') p_i1 = conditional_prob(counts, '1') p_i0_list.append(p_i0); p_i1_list.append(p_i1) print(f"[Z] phi={phi:.3f} P(s=0|i=0)={p_i0:.3f} P(s=0|i=1)={p_i1:.3f}") if save_mem: save_memory(memory, os.path.join(outdir, f"memory_Z_phi_{idx:02d}.txt")) save_csv_prob(phi_list, p_i0_list, p_i1_list, os.path.join(outdir, "whichpath_probs.csv")) print(f"Vizibilitate Z V(i=0)≈{visibility(p_i0_list):.3f} V(i=1)≈{visibility(p_i1_list):.3f}") if do_plot: plot_curves(phi_list, p_i0_list, p_i1_list, "Which-path (I in Z)", os.path.join(outdir, "whichpath_plot.png")) # ----------------------------- Main ------------------------------ def main(): ap = argparse.ArgumentParser(description="Delayed-choice quantum eraser (Qiskit + Aer)") ap.add_argument("--shots", type=int, default=4000, help="Număr de măsurători per punct φ (default: 4000)") ap.add_argument("--nphi", type=int, default=13, help="Număr de puncte între 0 și 2π (default: 13)") ap.add_argument("--outdir", type=str, default="out", help="Director ieșire (CSV, PNG, memorii)") ap.add_argument("--seed", type=int, default=1234, help="Seed pentru simulator (reproducibilitate)") ap.add_argument("--plot", action="store_true", help="Generează grafice PNG (matplotlib)") ap.add_argument("--memory", action="store_true", help="Salvează bitstring-urile brute pentru fiecare φ") args = ap.parse_args() os.makedirs(args.outdir, exist_ok=True) run_scan(nphi=args.nphi, shots=args.shots, outdir=args.outdir, seed=args.seed, save_mem=args.memory, do_plot=args.plot) print(f"Fișiere salvate în: {args.outdir}") if __name__ == "__main__": main()