Abstract: In an underdetermined system of equations , where is an matrix, only of the entries of with are known. Thus , called `measurements', are known for certain where are the rows of and . It is required, if possible, to solve the system uniquely when has at most non-zero entries with . Here such systems are considered from an error-correcting coding point of view. The unknown can be shown to be the error vector of a code subject to certain conditions on the rows of the matrix . This reduces the problem to finding a suitable decoding algorithm which then finds . Decoding workable algorithms are shown to exist, from which the unknown may be determined, in cases where the known values are evenly spaced (that is, when the elements of are in arithmetic progression) for classes of matrices satisfying certain row properties. These cases include Fourier matrices where the arithmetic difference satisfies , and classes of Vandermonde matrices (with ) with arithmetic difference where the ratios for are not roots of unity. The decoding algorithm has complexity and in some cases, including the Fourier matrix cases, the complexity is . Matrices which have the property that the determinant of any square submatrix is non-zero are of particular interest. Randomly choosing rows of such matrices can then give error-correcting pairs to generate a `measuring' code with a decoding algorithm which finds . This has applications to signal processing and compressed sensing.
