In the abstract, and the fourth and last paragraphs of [1], the phrase “20 percent faster than the Nyquist rate” contains a mistake. It should be corrected to “25 percent faster than the Nyquist rate.” As shown in our experiment, when $\alpha$ is set to 0.4, the baseband bandwidth is 4 GHz, and the corresponding Nyquist rate is 8 Gbit/s. However, the transmission rate (i.e., link rate) is 10 Gbit/s; therefore the transmission rate is 25% (i.e., $(10-8)/8=25\%$) faster than the Nyquist rate.

In the 18th paragraph, “the Nyquist the frequency of the signal with a 10 GS/s sample rate” (i.e., the ninth line of the fourth page) should be “the Nyquist frequency of the signal with a 10 GS/s sample rate.”