1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
use crate::noise_model::NoiseModel;
use crate::{SparseBinMat, SparseBinSlice, SparseBinVec};
use itertools::Itertools;
use rand::Rng;

mod random;
pub use self::random::RandomRegularCode;

/// An implementation of linear codes optimized for LDPC codes.
///
/// A code can be define from either a parity check matrix `H`
/// or a generator matrix `G`.
/// These matrices have the property that `H G^T = 0`.
///
/// # Example
///
/// This is example shows 2 way to define the Hamming code.
///
/// ```
/// # use ldpc::{LinearCode, SparseBinMat};
/// let parity_check_matrix = SparseBinMat::new(
///     7,
///     vec![vec![0, 1, 2, 4], vec![0, 1, 3, 5], vec![0, 2, 3, 6]]
/// );
/// let generator_matrix = SparseBinMat::new(
///     7,
///     vec![vec![0, 4, 5, 6], vec![1, 4, 5], vec![2, 4, 6], vec![3, 5, 6]]
/// );
///
/// let code_from_parity = LinearCode::from_parity_check_matrix(parity_check_matrix);
/// let code_from_generator = LinearCode::from_generator_matrix(generator_matrix);
///
/// assert!(code_from_parity.has_the_same_codespace_as(&code_from_generator));
/// ```
///
/// # Comparison
///
/// Use the `==` if you want to know if 2 codes
/// have exactly the same parity check matrix and
/// generator matrix.
/// However, since there is freedom in the choice of
/// parity check matrix and generator matrix for the same code,
/// use [`has_the_same_codespace_as`](LinearCode::has_the_same_codespace_as) method
/// if you want to know if 2 codes define the same codespace even
/// if they may have different parity check matrix or generator matrix.
#[derive(Debug, PartialEq, Eq, Clone, Hash)]
pub struct LinearCode {
    parity_check_matrix: SparseBinMat,
    generator_matrix: SparseBinMat,
}

impl LinearCode {
    /// Creates a new linear code from the given parity check matrix.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::{LinearCode, SparseBinMat};
    /// // 3 bits repetition code.
    /// let matrix = SparseBinMat::new(3, vec![vec![0, 1], vec![1, 2]]);
    /// let code = LinearCode::from_parity_check_matrix(matrix);
    ///
    /// assert_eq!(code.block_size(), 3);
    /// assert_eq!(code.dimension(), 1);
    /// assert_eq!(code.minimal_distance(), Some(3));
    /// ```
    pub fn from_parity_check_matrix(matrix: SparseBinMat) -> Self {
        Self {
            generator_matrix: matrix.nullspace(),
            parity_check_matrix: matrix,
        }
    }

    /// Creates a new linear code from the given generator matrix.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::{LinearCode, SparseBinMat};
    /// // 3 bits repetition code.
    /// let matrix = SparseBinMat::new(3, vec![vec![0, 1, 2]]);
    /// let code = LinearCode::from_generator_matrix(matrix);
    ///
    /// assert_eq!(code.block_size(), 3);
    /// assert_eq!(code.dimension(), 1);
    /// assert_eq!(code.minimal_distance(), Some(3));
    /// ```
    pub fn from_generator_matrix(matrix: SparseBinMat) -> Self {
        Self {
            parity_check_matrix: matrix.nullspace(),
            generator_matrix: matrix,
        }
    }

    /// Returns a builder for random LDPC codes with
    /// regular parity check matrix.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::LinearCode;
    /// use rand::thread_rng;
    ///
    /// let code = LinearCode::random_regular_code()
    ///     .block_size(20)
    ///     .number_of_checks(15)
    ///     .bit_degree(3)
    ///     .check_degree(4)
    ///     .sample_with(&mut thread_rng());
    ///
    /// assert_eq!(code.block_size(), 20);
    /// assert_eq!(code.number_of_checks(), 15);
    /// assert_eq!(code.parity_check_matrix().number_of_ones(), 60);
    /// ```
    pub fn random_regular_code() -> RandomRegularCode {
        RandomRegularCode::default()
    }

    /// Returns the parity check matrix of the code.
    pub fn parity_check_matrix(&self) -> &SparseBinMat {
        &self.parity_check_matrix
    }

    /// Returns the generator matrix of the code.
    pub fn generator_matrix(&self) -> &SparseBinMat {
        &self.generator_matrix
    }

    /// Checks if two code define the same codespace.
    ///
    /// Two codes have the same codespace if all there codewords are the same.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::{LinearCode, SparseBinMat};
    /// // The Hamming code
    /// let parity_check_matrix = SparseBinMat::new(
    ///     7,
    ///     vec![vec![0, 1, 2, 4], vec![0, 1, 3, 5], vec![0, 2, 3, 6]]
    /// );
    /// let hamming_code = LinearCode::from_parity_check_matrix(parity_check_matrix);
    ///
    /// // Same but with the add the first check to the other two.
    /// let parity_check_matrix = SparseBinMat::new(
    ///     7,
    ///     vec![vec![0, 1, 2, 4], vec![2, 3, 4, 5], vec![1, 3, 4, 6]]
    /// );
    /// let other_hamming_code = LinearCode::from_parity_check_matrix(parity_check_matrix);
    ///
    /// assert!(hamming_code.has_the_same_codespace_as(&other_hamming_code));
    /// ```
    pub fn has_the_same_codespace_as(&self, other: &Self) -> bool {
        (&self.parity_check_matrix * &other.generator_matrix.transposed()).is_zero()
    }

    /// Returns the number of bits in the code.
    pub fn block_size(&self) -> usize {
        self.parity_check_matrix.number_of_columns()
    }

    /// Returns the number of rows of the parity check matrix
    /// of the code.
    pub fn number_of_checks(&self) -> usize {
        self.parity_check_matrix.number_of_rows()
    }

    /// Returns the number of rows of the generator matrix
    /// of the code.
    pub fn number_of_generators(&self) -> usize {
        self.generator_matrix.number_of_rows()
    }

    /// Returns the number of linearly independent codewords.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::{LinearCode, SparseBinMat};
    /// let parity_check_matrix = SparseBinMat::new(
    ///     7,
    ///     vec![vec![0, 1, 2, 4], vec![0, 1, 3, 5], vec![0, 2, 3, 6]]
    /// );
    /// let hamming_code = LinearCode::from_parity_check_matrix(parity_check_matrix);
    ///
    /// assert_eq!(hamming_code.dimension(), 4);
    /// ```
    pub fn dimension(&self) -> usize {
        self.generator_matrix.rank()
    }

    /// Returns the weight of the smallest non trivial codeword
    /// or None if the code have no codeword.
    ///
    /// # Warning
    ///
    /// The execution time of this method scale exponentially with the
    /// dimension of the code.
    pub fn minimal_distance(&self) -> Option<usize> {
        (1..=self.number_of_generators())
            .flat_map(|n| self.generator_matrix.rows().combinations(n))
            .filter_map(|generators| {
                let weight = generators
                    .into_iter()
                    .fold(SparseBinVec::zeros(self.block_size()), |sum, generator| {
                        &sum + &generator
                    })
                    .weight();
                if weight > 0 {
                    Some(weight)
                } else {
                    None
                }
            })
            .min()
    }

    /// Returns the product of the parity check matrix with the given error.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::{LinearCode, SparseBinMat, SparseBinVec};
    /// let parity_check_matrix = SparseBinMat::new(
    ///     7,
    ///     vec![vec![0, 1, 2, 4], vec![0, 1, 3, 5], vec![0, 2, 3, 6]]
    /// );
    /// let hamming_code = LinearCode::from_parity_check_matrix(parity_check_matrix);
    ///
    /// let error = SparseBinVec::new(7, vec![0, 2, 4]);
    /// let syndrome = SparseBinVec::new(3, vec![0, 1]);
    ///
    /// assert_eq!(hamming_code.syndrome_of(&error.as_view()), syndrome);
    /// ```
    pub fn syndrome_of(&self, error: &SparseBinSlice) -> SparseBinVec {
        if error.len() != self.block_size() {
            panic!(
                "error of length {} is invalid for code of size {}",
                error.len(),
                self.block_size()
            );
        }
        self.parity_check_matrix.dot_with(error)
    }

    /// Checks if an operator has zero syndrome.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::{LinearCode, SparseBinMat, SparseBinVec};
    /// let parity_check_matrix = SparseBinMat::new(
    ///     7,
    ///     vec![vec![0, 1, 2, 4], vec![0, 1, 3, 5], vec![0, 2, 3, 6]]
    /// );
    /// let hamming_code = LinearCode::from_parity_check_matrix(parity_check_matrix);
    ///
    /// let error = SparseBinVec::new(7, vec![0, 2, 4]);
    /// let codeword = SparseBinVec::new(7, vec![2, 3, 4, 5]);
    ///
    /// assert_eq!(hamming_code.has_codeword(&error.as_view()), false);
    /// assert_eq!(hamming_code.has_codeword(&codeword.as_view()), true);
    /// ```
    pub fn has_codeword(&self, operator: &SparseBinSlice) -> bool {
        self.syndrome_of(operator).is_zero()
    }

    /// Generates a random error with the given noise model.
    ///
    /// # Example
    ///
    /// ```
    /// # use ldpc::{SparseBinMat, LinearCode};
    /// use ldpc::noise_model::BinarySymmetricChannel;
    /// use rand::thread_rng;
    ///
    /// let parity_check_matrix = SparseBinMat::new(
    ///     7,
    ///     vec![vec![0, 1, 2, 4], vec![0, 1, 3, 5], vec![0, 2, 3, 6]]
    /// );
    /// let code = LinearCode::from_parity_check_matrix(parity_check_matrix);
    ///
    /// let noise = BinarySymmetricChannel::with_probability(0.25);
    /// let error = code.random_error(&noise, &mut thread_rng());
    ///
    /// assert_eq!(error.len(), 7);
    /// ```
    pub fn random_error<N, R>(&self, noise_model: &N, rng: &mut R) -> SparseBinVec
    where
        N: NoiseModel<Error = SparseBinVec>,
        R: Rng,
    {
        noise_model.sample_error_of_length(self.block_size(), rng)
    }
}