add reference to analytic distribution

Author: hkwang

1_Dataset_prep_and_local_scaling.ipynb

@@ -33,12 +33,14 @@
     "    - $r = \\textbf{D}$ (or $\\sigma_E$ in eqs. 28-30).\n",
     "    - his $\\textbf{s}$ equal to the reciprocal lattice point vector which we calculate below as ```(rs_a, rs_b, rs_c)```.\n",
     "    - his $\\sigma^2_{\\Delta}$ related to our conditional variance $\\frac{1}{2}(1-r^2)$.\n",
-    " <br/><br/>      \n",
+    " <br/><br/>\n",
+    "\n",
+    "2. It is worth noting the closed-form relationship $r$ and the Pearson correlation has been derived in the signal processing literature, e.g. [(Tang and Kassam 2007)](https://digital-library.theiet.org/doi/abs/10.1049/iet-com%3A20050506), equation 20. \n",
     "     \n",
-    "2. The parameter $r$ used here is, in general, a function of resolution. Read's work (p. 903 in Read (1990)) addresses this at length and suggests that a Luzzati model ($r=a\\cdot e^{-b s^2}$) can be a good approximation.\n",
+    "3. The parameter $r$ used here is, in general, a function of resolution. Read's work (p. 903 in Read (1990)) addresses this at length and suggests that a Luzzati model ($r=a\\cdot e^{-b s^2}$) can be a good approximation.\n",
     " <br/><br/>\n",
     " \n",
-    "3. notation:\n",
+    "4. notation:\n",
     "    - we will use $hkl$ and ```HKL```\n",
     "    - we will denote reciprocal lattice point coordinates as ```(rs_a,rs_b,rs_c)``` and $r^*$, with magnitude $1/d_{hkl}$ or ```dHKL```. The scattering vector $s=S_1-S_0$ equals $r^*$ in our case (elastic scattering)."
    ]