### Table 1. The true and estimated poles of the transfer function H(z)

1996

"... In PAGE 3: ... Once b A(z) is obtained, B(z) can be estimated us- ing Shanks apos; method [8], which rst generates a sequence fn by f(n) = Z?1f 1 b A(z)g; (13) and then estimate bk for k = 0; 1; 2 by minimiz- ing the error E = 39 X n=0 j h(n) ? 2 X k=0 bbkf(n ? k) j2 : (14) Figure 2 (b), (c) and (d) show 10 trails of mag- nitudes of estimated transfer function using KT, MKT and the new matrix pencil algorithms re- spectively. Also Table1 illustrates the mean and variance of the estimated poles of transfer func- tion using KT, MKT and the new matrix pencil algorithms. From Figure 2 and Table 1, it is clear that the new matrix pencil algorithm gives more accurate estimate of transfer function H(z) than KT and MKT algorithms.... In PAGE 3: ... Also Table 1 illustrates the mean and variance of the estimated poles of transfer func- tion using KT, MKT and the new matrix pencil algorithms. From Figure 2 and Table1 , it is clear that the new matrix pencil algorithm gives more accurate estimate of transfer function H(z) than KT and MKT algorithms. 5.... ..."

Cited by 3

### Table 2: The true and estimated poles of the transfer function H(z)

1998

"... In PAGE 13: ... Figures 3 (b), (c) and (d) show the results. Table2 shows the mean and variance of the estimated poles of the transfer function using KT, MKT and the new matrix pencil algorithms. From Figure 3 and Table 2, it is clear that the new matrix pencil algorithm outperforms the KT and MKT algorithms.... In PAGE 13: ...nd the new matrix pencil algorithms respectively. Figures 3 (b), (c) and (d) show the results. Table 2 shows the mean and variance of the estimated poles of the transfer function using KT, MKT and the new matrix pencil algorithms. From Figure 3 and Table2 , it is clear that the new matrix pencil algorithm outperforms the KT and MKT algorithms. 5 Conclusion In this paper a new matrix pencil algorithm for estimating the parameters (frequencies and damping factors) of exponentially damped sinusoids in noise is proposed.... ..."

Cited by 4

### Table 4: Computational complexity requirement to implement the Full Bayesian equaliser and subset equaliser using m = 4 for channel H(z).

"... In PAGE 4: ... The results show that when is set to values greater than 4 n, the number of exceptions which occurs becomes very small. Table4 compares the computational requirement between the full RBF Bayesian equaliser and the subset equaliser. The results clearly show that the implementation complexity of the subset equaliser is much lower than that of the full RBF equaliser.... ..."

### Table1. Values of the bit-error rate obtained by equalisers for the channel H(z) = 1 + 0:7z?1 over 30 runs.

"... In PAGE 6: ... It is therefore interesting to try and nd low order equalisers which employ some form of nonlinearity. Results for di erent values of the SNR are given in Table1 and shown graph- ically in Fig 2. Average and minimum values of the BER for 30 runs (per point) are presented, showing that the GP+SA method can obtain lower minimum values than the FIR-RLS, especially for values of the SNR of 7.... ..."