# Quick Performance Assessment of Improved Nyquist Pulses.

1. IntroductionIn the design of a Nyquist filter the robustness to timing jitter is a prime factor. A solution is to redistribute the tail energy of impulse response by diminishing the size of the largest sidelobe and in turn increase the size of the subsequent sidelobes. The result is a decrease of the intersymbol interference (ISI), which manifests itself in a decrease of the error rate when the impulse response is sampled at the receiver's site with a timing error. This approach was first proposed and used in [1].

Beaulieau et al. [1, 2] demonstrated that a Nyquist pulse that decays asymptotically as [t.sup.-2] performs better than a pulse with an asymptotic decay rate (ADR) of [t.sup.-3], such as the standard raised cosine (RC or rcos) pulse, in terms of error probability when sampled with a timing offset. In practice the pulse is generated digitally using a truncated version of the impulse response of the Nyquist filter and is digitally filtered at the receiver, using an adapted filter to its truncated version, which introduces spectral re-growth.

The new pulses with a slower ADR ([t.sup.-2] versus [t.sup.-3]) require a bigger number of taps when implemented digitally in a finite impulse response (FIR) structure, as compared with the RC pulse. Stated in another way, for a given truncated length, the spectral regrowth of a slowly decaying pulse is worse than that of a fast decaying pulse. In order to obtain comparable performance in terms of spectral regrowth, one should increase the number of filter taps, resulting in bigger latency and implementation costs.

In some applications where the latency is not critical these disadvantages are counteracted by the fact that one obtains increased performance in terms of lower bit error rate (BER).

Several researchers embraced this idea known as improved Nyquist filter (INF) or pulse and produced on heuristic bases novel pulses that performed better than the ones previously reported [3-11]. However, the mechanism behind the improved performance of Nyquist pulses was not completely understood so far. In the sequel we investigated the behavior of several improved Nyquist pulses that decay asymptotically as [t.sup.-2] and [t.sup.-3] and proposed three figures of merit to quickly assess their performance in terms of error probability when the impulse response is sampled with a timing error.

Also, a novel family of Nyquist pulses, denoted as power sine was introduced in order to check the validity of the approach to expedite the design process.

2. Fractional Energy

To m[a.sub,k]e the tails of the impulse response less damped, that is, to increase its oscillatory feature it is necessary to transfer an amount of energy from the lower frequency part of the transition region [B(1-[alpha]),B] into the higher one [B, B(1+[alpha])] [7, 9]. The parameter [alpha] is known as excess bandwidth or rolloff factor.

his is to say the frequency characteristic should be concave in the [B(1 - [alpha], B] range and in view of odd-symmetry convex in the [B, B(1 + [alpha])] range. Otherwise stated, we wish to decrease the fractional energy contained in the lower frequency band [B(1 - [alpha]), B] of the time-unlimited Nyquist pulse by transferring some energy into the higher frequency band [B, B(1 + [alpha])], that is, to increase the fractional energy contained in the frequency band [B, B(1 + [alpha])] up to some level.

As a figure of merit regarding the efficiency of this transfer we propose to take the ratio of the pulse energy contained either in the frequency interval [B, B(1 + [alpha])] or [B(1 - [alpha]),B] to the total energy contained in the frequency interval [0, B(1 + [alpha])], by taking into account only the positive frequency components. We will denote it as fractional energy [E.sub.s] and [E.sub.l], respectively,

[mathematical expression not reproducible] (1)

Also,

[mathematical expression not reproducible] (2)

For a Nyquist filter the energy carried in the frequency interval [0, [beta] (1 - [alpha])] is obviously (1 - [alpha]) for B =1. So,

[E.sub.l] + [E.sub.s] = [alpha]. (3)

This is justified by the fact that H(f) has unit energy, as this is a Nyquist filter.

A higher value of the fractional energy [E.sub.s] denotes that more energy is transferred within the transition band of the filter from [B(1 - [alpha]),B] to the [B, B(1 + [alpha])] frequency interval.

A similar approach was found in [12] where a similar metric was used. This is calculated by taking the ratio of the root-Nyquist pulse energy between 1/(2T) and infinity, to its total energy.

Table 1 reports the analytic expressions of fractional energy for several Nyquist pulses that are mathematically tractable. For other pulses one must use numerical integration for given values of excess bandwidth [alpha].

3. Slope of the Time Response at Its First Zero Crossing

The improved performance of the new Nyquist pulse in [1] when sampled with a time offset is explained based on the fact that the magnitudes of the two largest sidelobes [1] are smaller than the magnitudes of the RC pulse. Also in [7] one shows that the new Nyquist pulses exhibit a more pronounced decrease in the amplitudes of the two largest sidelobes, which accounts for their improved robustness to error probabilities.

Based on these assertions two figures of merit are proposed.

The first one is the slope of the time response evaluated at its first zero crossing, t/T = 1. Obviously, the higher the slope of time response evaluated at t/T = 1, the higher the magnitude of the first sidelobe is expected to be.

Several pulses, such as RC, bTrC (fexp) [1], Poly [5], and Power [8], have time responses given by closed-form formulae, which allows to obtain the first derivative of the time response as a closed-form expression. Table 2 reports the analytic expressions of the slope of the time response evaluated at t/T = 1 for the pulses mentioned above, where x = [alpha][pi], and [sup.1][F.sub.2][{a},{b,c], y] is a hypergeometric function [8] given by

[mathematical expression not reproducible] (4)

with [(z).sub.0] = 1 and [(z).sub.k] = z(z + 1)(z + 2) ... (z + k - 1).

For other pulses their impulse response is evaluated through a numerical inverse Fourier transform [3]. In this case one can use the slope interpretation to gain rough information about the derivative, which is the slope of the tangent line at the considered point.

In practice one often approximates the derivative by a difference quotient. After performing the numerical inverse Fourier transform the impulse response h(t) is obtained as a set of discrete values {[h.sub.k]} at equally spaced time values given by the set {[t.sub.k]}. If the sampling frequency was chosen high enough, say 100 samples per symbol interval, the points {[h.sub.k]} in the table are close enough together.

Given a table of values for the time response one can estimate the values of its derivative. So, the function does not change abruptly between them and the derivative is closely approximated by the difference quotient [s.sub.k] defined as

[s.sub.k] = [h.sub.k + 1] - [h.sub.k-1]/[t.sub.k + 1] - [t.sub.k - 1] (5)

where [h.sub.k] is the sample of the impulse response at t/T = 1, [h.sub.k+1] and [h.sub.k-1] are the samples adjacent to [h.sub.k].

The pulses reported in Table 2 that have a derivative expressed by a closed-form formula have been used for checking the accuracy of the calculations based on the difference quotient. Rel. (5) gave the best match.

4. Second Derivative of the Time Response Evaluated at t/T = 1

We propose also to use the second derivative of the time response evaluated at t/T = 1 as a figure of merit related to the size of the largest sidelobe. A bigger value of acceleration combined with a negative value of velocity, which is the case for the descending part of the largest sidelobe, will lead to a faster decrease of the time response. This results in increased performance when the time response is sampled with a timing error.

The second derivative of the impulse response was calculated for four pulses that give closed-form expressions of the time response and reported in Table 3, where x = [alpha][phi], A = 120d + 12(c - 2b)[x.sub.2] + (8 + 4b + 2c + d)[x.sup.4], B = 2x(24c-12d) + (8 + 8b + 2c + d)[x.sup.2], C = [2.sup.-{2+[beta])][[pi].sup.5/2][[alpha].sup.2]T[1 + [beta]], [sup.1][F.sub.2][{a}, {b, c}, y] is defined by rel. (4), and [GAMMA][t] is the Euler gamma function:

[mathematical expression not reproducible] (6)

For other pulses that do not have time responses expressed by closed-form formulae the derivative of the slope [s.sub.k] at t/T = 1 was obtained numerically. The value of the derivative of the slope [s.sub.k] at the time instant [t.sub.k], which can be envisaged as acceleration, was calculated as

[a.sub.k] = [s.sub.k+1] - [s.sub.k-1]/[t.sub.k+1] - [t.sub.k-1]. (7)

Comparing the values of the difference quotient [a.sub.k] with the values of the second derivative of the pulses determined with the closed-form formulae reported in Table 3 and evaluated at t/T = 1, almost perfect match was found.

5. Comparative Analysis of Several Improved Pulses

In the sequel we have considered several improved pulses reported so far, such as RC, BTRC [1] also known as fexp [3], fsech [3], farcsech [3], Poly [5], acos [7], acos[acos] [7], acos[asech] [7], acos[log] [7], asech[acos] [7], asech[asech] [7], asech[log] [7], asech [exp] [7], acos[exp] [7], Power [8], acos[asinh] [6], acos[atan] [6], and sin[acosh] [6].

We have proceeded to rank the above mentioned pulses with respect to the error probability in descending order. We have correlated the results with the fractional energy defined as above or calculated by numerical integration.

Tables 4, 5, 6, and 7 report the values of error probability, fractional energy, and first and second derivatives of time response evaluated at t/T = 1, for several cases of practical interest, namely, [alpha] = 0.1, [alpha] = 0.25, [alpha] = 0.35, and [alpha] = 0.5, respectively, and three values of the normalized time offset, 0.05, 0.1, and 0.2. For the s[a.sub,k]e of compactness we have denoted the normalized time offset t/T as [epsilon].

Special attention must be paid to Poly and Power pulses, as they do not have a fixed form for a given value of [alpha], as the other pulses, but depend on one or three parameters, respectively. For the s[a.sub,k]e of compactness one usually adopted a set of truncated values determined for [epsilon] = 0.1 and rounded to the next integer, which may lead to differences in performance.

The pulses have been placed in ascending order of performance with respect to error probability. We have found that for small values of the normalized time offset [epsilon] = 0.05, [epsilon] = 0.1, and [epsilon] = 0.2, the pulses reported in ascending order of performance with respect to error probability are also placed in ascending order of fractional energy [E.sub.s], descending order of the absolute value of the slope [s.sub.k] of the impulse response evaluated at t/T = 1, and ascending order of the second derivative of the impulse response [a.sub.k] evaluated at t/T = 1.

However, for [alpha] = 0.25, and [epsilon] [greater than or equal to] 0.1 there are two exceptions in Table 5 marked in bold characters, involving the asech[asech] and asech[acos] pulses with close values of fractional energy [E.sub.s] of 0.0877, and 0.0880, respectively. The values of error probability evaluated for [epsilon]= 0.1 and [omega] = 0.2 do not obey the ranking order given by the fractional energy Es, the slope value [s.sub.k] at t/T = 1, and acceleration [a.sub,k] at t/T = 1. Despite its higher values of [E.sub.s], [s.sub.k], and [a.sub.k], the asech[acos] pulse is slightly outperformed by the asech[asech] pulse for timing offsets of 0.1 and 0.2.

The bold characters mark the situation that despite better values of [E.sub.s], [s.sub.k], and [a.sub.k], the value of error probability of asech[acos] pulse for [epsilon] [greater than or equal to] 0.1 is worse with respect to the asech[asech] pulse, while for [epsilon] = 0.05 this correlation between error probability and the [E.sub.s], [s.sub.k], and [a.sub,k] values holds. Although the difference in error probability for [epsilon] = 0.2 is quite small, around 1.4%, this deserves further analysis.

6. Further Analysis

It is worth mentioning that the acos[acos], asech[exp], asech[acos], and asech[asech] pulses have almost identical frequency characteristics, which results in very close values of the error probability, the difference in error probability values being under 0.05% for small values of the normalized time offset.

The difference in the values of the main lobe sample h(e) is very small, so to explain the difference in performance we have introduced the limited ISI distortion for a time offset [epsilon], produced by k interfering pulses denoted as [d.sup.[epsilon].sub.k], and defined as

[d.sup.[epsilon].sub.k] = [k.summation over (i=1)] [absolute value of ]h(i + [epsilon])]. (8)

The absolute value of the sample of the nth lobe for a specified value of [epsilon] can be found also as

[absolute value of ]h(n + [epsilon])]= [d.sup.[epsilon].sub.n] - [d.sup.[epsilon].sub.n-1]. (9)

For instance [absolute value of ]h(4.1)] for asech[asech] pulse is found as [absolute value of ]h(4.1)] = [d.sup.0.1.sub.4] - [d.sup.0.1.sub.3 ]= 0.1034 - 0.0920 = 0.0114, while for asech[acos] pulse [absolute value of ]h(4.1)] = 0.0115.

To understand what determines the difference in performance for pulses with almost the same value of the error probability and fractional energy, we have calculated the values of the main lobe sample h([epsilon]) and of [d.sup.[epsilon].sub.k], with [epsilon] = 0.1 and [epsilon] = 0.2, for a pair of pulses with almost equal performance in terms of error probability, namely, asech[asech] and asech[acos] that do not obey the proposed ranking criteria and also for the Poly pulse that shows a greater value of error probability, as expected based on fractional energy value for [epsilon] = 0.2. This is shown in Table 8 that presents a comparative analysis of asech[asech], and asech[acos] and Poly pulses for the excess bandwidth value [alpha] = 0.25. The entries in bold characters specify the advantages of a pulse with respect to the other one in a set of pulses with almost the same value of the error probability and fractional energy.

Sampling the Nyquist pulse at the receiver's site with a time offset [epsilon], that is, sampling off-pe[a.sub,k] incurs a penalty, which can be seen in the value of h([epsilon]). For [epsilon] = 0.2 the samples are taken far from the center of the eye and the value of h[epsilon] is decreased.

There are two factors that contribute to the error mechanism. One is the decreased value of the main sample h([epsilon]) and the other one is the residual effect of all other transmitted bits, known as ISI. The improved pulses have an asymptotic decay rate of [t.sup.-2] when sampled with a timing offset. So, the higher order sidelobes are less important than the low order sidelobes in determining the error performance.

An analysis of the pulses reported in Table 8 reveals that asech[asech] and asech[acos] pulses have almost equal performance. As the sample of the main lobe h[epsilon] is almost the same, the small difference in performance may be attributed to the difference in the absolute values of the sample of the low order sidelobes, say 3th to 10th.

For [epsilon] = 0.2 the asech[asech] pulse has an advantage also due to the slightly higher value of h(0.2).

Although the asech[acos] pulse has a smaller value of [d.sup.0.2.sub.2] than asech[asech] pulse and the same value of [d.sup.0.2.sub.1] its performance is worse due to the higher values of [d.sup.0.2.sub.k], k > 3. A higher value of [d.sup.[epsilon].sub.k] evidences that, for the specified time offset [epsilon], the sum of absolute values of the samples of the sidelobes up to kth sidelobe is larger and produces more ISI that results in increased error probability value.

The improved behavior is explained in [2] by examination of the inner sidelobes of the pulses. Improved performance is obtained if the magnitudes of the two largest sidelobes of the pulse under consideration are smaller as compared with the magnitudes of the two largest sidelobes of the raisedcosine pulse [1]. Also the maximum distortion is less for the improved pulses than the raised-cosine pulse [2].

When dealing with improved pulses of almost equal performance this is not enough and one must also consider the magnitudes of the samples of the low order sidelobes, up to the 10th side lobe.

For [alpha] = 0.35 several exceptions regarding the order marked in bold characters were evidenced, involving acos[acos], acos, asech[exp], asech[acos], and asech[asech] pulses.

In Table 9 we have considered the asech[asech], asech[acos], acos[asech], acos, acos[acos], and asech[exp] pulses for the cases when they do not obey the ranking rule based on fractional energy and values of the first and second derivative evaluated at t/T = 1.

For [alpha] = 0.35 and [epsilon] = 0.1, despite having a larger value of [E.sub.s], the asech[acos] pulse is slightly outperformed by the asech[asech] pulse. This can be attributed to the larger value of [d.sup.[epsilon].sub.k], 3 [less than or equal to] k [less than or equal to] 10 for asech[acos] pulse.

For small values of timing offset, [epsilon] [less than or equal to] 0.1, the fractional energy criterion is prevalent. For higher values of timing offset and close values of the fractional energy one should also consider the limited ISI distortion [d.sup.[epsilon].sub.k] with k up to 10.

The values of error probability, fractional energy, slope [s.sub.k], and acceleration [a.sub,k] are tabulated in Table 7 for [alpha] = 0.5.

Table 10 presents a comparative analysis of several pairs of pulses with close values of fractional energy from Table 7 for the cases when they do not obey the proposed ranking criteria.

The advantage of a pulse over the other one in the same pair was marked in bold characters, that is, higher values of Es and main lobe sample on one hand and lower absolute values of the samples of sidelobes up to the kth one on the other hand.

As inferred from Table 10, for close values of fractional energy, the pulse with a smaller value of the fractional energy outperforms the other pulse in the pair for higher values of the timing offset due to a decrease in the magnitude of the side lobes up to the kth one, with k ranging in the interval [4,10] and a slight increase in the magnitude of the main lobe sample for some pulses.

7. Power Sine Pulses

In order to check that the fractional energy can be applied to the design of an improved Nyquist pulse, a novel family of pulses known as power sine (PS) was proposed. The PS pulse shows a frequency characteristic that is concave in the frequency range [B(1 - [alpha]), [epsilon]]. So, an energy transfer is made from the lower frequency range [B(1 - [alpha]),B] to the higher one [B, B(1 + [alpha])] in order of the odd symmetry of the Nyquist frequency characteristic. As a result, improved performance is expected as compared with the RC pulse [7, 9].

Its frequency spectrum is defined as

[mathematical expression not reproducible] (10)

Figure 1 illustrates the frequency characteristic of PS pulse for [alpha] = 0.25 and n = 0.25, 0.5, and 1.

The design of an improved Nyquist pulse with an impulse response that depends on several parameters and not given by a closed-form expression is a time-consuming process. The impulse response must be evaluated through a numerical inverse Fourier transform for a given set of parameters and then one calculates the error probability when the impulse response is sampled with a time offset [epsilon]. If [epsilon] = 0.05 this implies the use of 20 samples per symbol interval.

Usually the average symbol error probabilities of binary antipodal signaling [12] in the presence of symbol timing error implies the use of 512 interfering symbols, so the impulse response is described by a vector with 10240 components. The process must be repeated many times using different values of parameters. Consider the PS pulse. The parameters involved in the calculation of the error probability of PS pulse are the excess bandwidth a and the exponent n.

In order to expedite the design process of a PS filter with good performance for a specified value of a, one should determine the values of n for which fractional energy is optimal.

In other cases, if [E.sub.s] cannot be obtained as a closed-form expression, one should use numerical integration and calculate it for a set of values.

The fractional energy of PS pulse was obtained as

[E.sub.s] = a/2 [GAMMA][(1 + n)/2]/[square root of [pi]][GAMMA] [1 + n/2]. (11)

In order to get improved performance, one should first obtain the impulse response using a numerical integration of the frequency characteristic for a set of values of n. This is a time-consuming operation followed by the calculation of the error probability and the comparison with the best pulses reported so far, for example, Poly [5] and Power [8].

To speed up the design process one should reduce the set size of n values.

Assume [alpha] = 0.1 and that we want to obtain a PS pulse that outperforms the Poly pulse. The fractional energy value of the optimal Poly pulse for [alpha] = 0.1 is 0.0437.

Figure 2 illustrates the variation of [E.sub.s] with n for the PS pulse for [alpha] = 0.1 and [alpha] = 0.25.

Assuming a 10% error margin, we need to determine the error probability for values of n around 0.25 and below it for [alpha] = 0.1. The results are presented in Table 11.

One can see that the PS pulse outperforms the Poly pulse for n [less than or equal to] 0.3 but it is outperformed by the Power pulse, despite its greater value of [E.sub.s]. As the error probability values are very close to each other, it is obvious that there are other factors to be considered.

Assume [alpha] = 0.25. The Poly pulse has [E.sub.s] = 0.0937. The PS pulse will have a maximum of performance if its fractional energy [E.sub.s] is close to this value. This reaches 0.09375 for n = 0.54. Assuming a 10% error margin, we need to determine the error probability for values of n around 0.6 and below it. The results are tabulated in Table 12.

For [alpha] = 0.25, [epsilon] = 0.05, n = 0.2, and n = 0.25 the PS pulse outperforms both Poly and Power pulses.

8. Conclusions

To explain the improved performance of Nyquist pulses in the presence of timing errors three figures of merit based on fractional energy, first and second derivative of the time response, in short velocity and acceleration, evaluated at first zero crossing of the impulse response have been proposed and verified on pulses that have been reported in the corresponding literature and on a novel family of pulses. Small discrepancies appear for higher values of the excess bandwidth and symbol timing errors. All three figures of merit indicate the energy distribution into the sidelobes of the time response that leads to improved performance when the impulse response is sampled with a time offset and are related to the pulse performance. The mechanism of energy distribution based on fractional energy, velocity, and acceleration, evaluated at first zero crossing of the impulse response, works very well for small and moderate values of excess bandwidth and time offset. So, all these criteria can be used for quick performance assessment of improved Nyquist pulses. For small values of excess bandwidth, say [alpha] = 0.1 and [alpha] = 0.25, the proposed criteria based on fractional energy and values of the first and second derivative of the time response evaluated at t/T = 1 are valid to rank the pulses based on error probability performance when sampled with usual values of the time offset, say 0.05, 0.1, and 0.2. Caution must be exerted when investigating pulses with close values of fractional energy, as exceptions may occur. Also the criteria based on the values of the first and second derivative of the time response evaluated at t/T = 1 (velocity and acceleration) must be used with caution taking into account that they are not independent. For instance, considering a pair of pulses with close values of fractional energy, velocity and acceleration, different results maybe obtained taking into account that the decrease of the magnitude of the largest side lobe can be produced by either a bigger value of velocity combined with a smaller value of acceleration or vice versa. Also, the proposed criteria for ranking the pulses should be used with caution when comparing pulses with different asymptotic decay rates at higher values of excess bandwidth a. For larger values of excess bandwidth a and higher values of the time offset, say t/T = 0.2, when the pulses have values of [E.sub.s] that are close to each other, the difference in error probability is no longer determined only by the magnitude of the two largest sidelobes [7], but also by the differences in the magnitude of the sidelobes up to the kth one, with 3 [less than or equal to] k [less than or equal to] 10. The quick assessment method proposed here was verified on a novel family of improved Nyquist pulses, denoted as power sine. This significantly restricted the value range of the design parameter n to be searched in order to obtain improved performance. To explain the difference in pulse performance a figure of merit based on the limited ISI distortion was introduced. The mechanism of performance improvement was not completely deciphered so far, as for each improved pulse there is an optimal value of [E.sub.s], [s.sub.k], and [a.sub,k] that once surpassed may lead to poorer performance.

http://dx.doi.org/ 10.1155/2017/7071648

Competing Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

References

[1] N. C. Beaulieu, C. C. Tan, and M. O. Damen, "A better than Nyquist pulse," IEEE Communications Letters, vol. 5, no. 9, pp. 367-368, 2001.

[2] N. C. Beaulieu and M. O. Damen, "Parametric construction of Nyquist-I pulses," IEEE Transactions on Communications, vol. 52, no. 12, pp. 2134-2142, 2004.

[3] A. Assalini and A. M. Tonello, "Improved nyquist pulses," IEEE Communications Letters, vol. 8, no. 2, pp. 87-89, 2004.

[4] P. Sandeep, S. Chandan, and A. K. Chaturvedi, "ISI-free pulses with reduced sensitivity to timing errors," IEEE Communications Letters, vol. 9, no. 4, pp. 292-294, 2005.

[5] S. Chandan, P. Sandeep, and A. K. Chaturvedi, "A family of ISI-free polynomial pulses," IEEE Communications Letters, vol. 9, no. 6, pp. 496-498, 2005.

[6] S. D. Assimonis, M. Matthaiou, and G. K. Karagiannidis, "Two-parameter Nyquist pulses with better performance," IEEE Communications Letters, vol. 12, no. 11, pp. 807-809, 2008.

[7] S. D. Assimonis, M. Matthaiou, G. K. Karagiannidis, and J. A. Nossek, "Improved parametric families of intersymbol interference-free Nyquist pulses using inner and outer functions," IET Signal Processing, vol. 5, no. 2, pp. 157-163, 2011.

[8] M. Mohri and M. Hamamura, "ISI-free power roll-off pulse," IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, vol. E92-A, no. 10, pp. 24952497, 2009.

[9] N. D. Alexandru and A. L. O. Balan, "Improved nyquist filters with piece-wise parabolic frequency characteristics," IEEE Communications Letters, vol. 15, no. 5, pp. 473-475, 2011.

[10] A. L. Balan and N. D. Alexandru, "Two improved nyquist filters with piece-wise rectangular-polynomial frequency characteristics," AEU--International Journal of Electronics and Communications, vol. 66, no. 11, pp. 880-883, 2012.

[11] N. D. Alexandru and A. L. Balan, "ISI-free pulses produced by improved Nyquist filter with piece-wise linear characteristic," Electronics Letters, vol. 47, no. 4, pp. 256-257, 2011.

[12] N. C. Beaulieu, "The evaluation of error probabilities for intersymbol and cochannel interference," IEEE Transactions on Communications, vol. 39, no. 12, pp. 1740-1749,1991.

Nicolae Dumitru Alexandra and Felix Diaconu

Department of Telecommunications, "Gheorghe Asachi" Technical University of Iasi, Bd. Carol I, No. 11A, 700506 Iasi, Romania

Correspondence should be addressed to Felix Diaconu; fdiaconu@etti.tuiasi.ro

Received 7 July 2016; Accepted 20 September 2016; Published 11 January 2017

Academic Editor: Javier Del Ser

Caption: Figure 1: Frequency characteristics of the PS pulse for an excess bandwidth [alpha] = 0.25 and n = 0.25, 0.5, and 1.

Caption: Figure 2: Variation of fractional energy of PS pulse and optimal Poly pulse for [alpha] = 0.1 and [alpha] = 0.25.

Table 1: Analytic expressions of fractional energy for several Nyquist pulses. Pulse [E.sub.s] rcos [alpha]B/2[pi] ([pi]-2) fsech [alpha]B/2[pi] (1 - [pi]/3 x a cosh (2)) fexp [alpha]B - [alpha]B/log (4) farcsech (B = 1) 2[pi][alpha]/3 - [alpha] log (2 + [squareroot of 3])/2 x a cosh (2) Power (B = 1) [alpha]/2(1 + [beta]) Poly ([alpha], b, c, d) [alpha][20b + 15(c + 8) + 9d]/480 Pulse [E.sub.l] rcos [alpha]B/2[pi] ([pi] + 2) fsech [alpha]B[pi]/3 x a cosh (2) fexp [alpha]B/log (4) farcsech (B = 1) [alpha] - 2[pi][alpha]/3 - [alpha] log (2 + [square root of 3])/2 x a cosh (2) Power (B = 1) [alpha] - [alpha]/2 (1 + [beta]) Poly ([alpha], b, c, d) [alpha][20b + 15c + 9(d - 40)]/480 Table 2: Analytic expressions of slope at t/T = 1 for several Nyquist pulses. Pulse Slope at t/T = 1 rcos -cos ([alpha][pi])/1 - 4[[alpha].sup.2] fexp ln (2) [ln (2)--cos ([alpha][pi]) ln (4)-- 2[alpha][pi] sin ([alpha][pi])] ([alpha][pi])2 + [[ln (2)].sup.2] Poly (b, c, d) 24d--4 [2b + 3 (c + d)] [x.sup.2] + 8 ([bx.sup.2]--3d) cos (x) + x [12c + (8 + 4d + 2c + d) [x.sup.2]] sin (x)/8[x.sup.4] Power 1 + [x.sup.2] [sub.1][F.sub.2], [{1}, {3/2 + [beta]/2,2 + [beta]/2}, - (1/4) [x.sup.2]] 2 + 3[beta] + [[beta].sup.2] Table 3: Analytic expressions of second derivative of time response at t/T = 1 for several Nyquist pulses. Pulse Derivative of slope at t/T = 1 rcos [alpha][pi](1 - 4[[alpha].sup.2]) sin ([alpha][pi]) + (1- 12[[alpha].sup.2]) cos ([alpha][pi])/ 1 - 4[[alpha].sup.2] fexp -2 ln(2) [[ln (2)].sup.3] + 2 cos (x) {[x.sup.4] + [x.sup.2] [ln (2) -3] ln (2)-[[ln (2)].sup.3]} + [x.sup.2] ln (8) -2x {[[[ln (2)].sup.3] + [x.sup.2] [2 + ln (2)]]} sin (x)/ [{[([alpha][pi]).sup.2] + [ln (2)]2}.sup.2] Poly (b, c, d) 120d - 12 [2b + 3 (c + d)] [x.sup.2] - A cos (x) + B sin (x)/4x4 Power [mathematical expression not reproducible] Table 4: ISI error probabilities of several Nyquist pulses for N = [2.sup.9] interfering symbols, [T.sub.f] = 40, M = 61, SNR =15 dB, and [alpha] = 0.1. Pulse [P.sub.e] [P.sub.e] [alpha] = 0.1 [epsilon] = 0.05 [epsilon] = 0.1 rcos 1.4130e - 7 9.2400e - 6 fsech 1.3583e - 7 8.4906e - 6 asech[log] 1.2292e - 7 6.8473e - 6 fexp 1.1926e - 7 6.4133e - 6 farcsech 1.1484e - 7 5.9066e - 6 acos[log] 1.1407e - 7 5.8221e - 6 acos[atan] 1.1381e - 7 5.7935e - 6 sin[acosh] 1.1360e - 7 5.7710e - 6 acos[asinh] 1.1183e - 7 5.5799e - 6 acos[exp] 1.1126e - 7 5.5190e - 6 acos[asech] 1.1083e - 7 5.4740e - 6 acos 1.1074e - 7 5.4647e - 6 acos[acos] 1.0917e - 7 5.3032e - 6 asech[exp] 1.0879e - 7 5.2647e - 6 asech[asech] 1.0806e - 7 5.1906e - 6 asech[acos] 1.0803e - 7 5.1883e - 6 Poly (60, -152,130) 1.0355e - 7 4.8248e - 6 Power ([beta] = 0.07) 1.0073e - 7 4.5696e - 6 Pulse [P.sub.e] [E.sub.s] [s.sub.k] [alpha] = 0.1 [epsilon] = 0.2 t/T = 1 rcos 3.3488e - 3 0.0182 -0.9907 fsech 3.1092e - 3 0.0205 -0.9888 asech[log] 2.5495e - 3 0.0245 -0.9828 fexp 2.3933e - 3 0.0279 -0.9808 farcsech 2.2068e - 3 0.0295 -0.9779 acos[log] 2.1750e - 3 0.0300 -0.9774 acos[atan] 2.1642e - 3 0.0303 -0.9772 sin[acosh] 2.1557e - 3 0.0305 -0.9770 acos[asinh] 2.0832e - 3 0.0318 -0.9755 acos[exp] 2.0599e - 3 0.0323 -0.9750 acos[asech] 2.0427e - 3 0.0325 -0.9746 acos 2.0391e - 3 0.0327 -0.9745 acos[acos] 1.9767e - 3 0.0341 -0.9730 asech[exp] 1.9617e - 3 0.0345 -0.9725 asech[asech] 1.9329e - 3 0.0351 -0.9718 asech[acos] 1.9320e - 3 0.0352 -0.9717 Poly (60, -152,130) 1.7843e - 3 0.0437 -0.9608 Power ([beta] = 0.07) 1.6840e - 3 0.0467 -0.9558 Pulse [a.sub.k] [alpha] = 0.1 t/T = 1 rcos 2.0185 fsech 2.0222 asech[log] 2.0340 fexp 2.0381 farcsech 2.0436 acos[log] 2.0447 acos[atan] 2.0451 sin[acosh] 2.0454 acos[asinh] 2.0483 acos[exp] 2.0493 acos[asech] 2.0500 acos 2.0503 acos[acos] 2.0534 asech[exp] 2.0542 asech[asech] 2.0558 asech[acos] 2.0559 Poly (60, -152,130) 2.0773 Power ([beta] = 0.07) 2.0869 Table 5: ISI error probabilities of several Nyquist pulses for N = [2.sup.9] interfering symbols, [T.sub.f] = 40, M = 61, SNR = 15 dB, and [alpha] = 0.25. Pulse [P.sub.e] [P.sub.e] [alpha] = 0.25 [epsilon] = 0.05 [epsilon] = 0.1 rcos 8.2189e - 8 2.8184e - 6 fsech 7.5579e - 8 2.3337e - 6 asech[log] 6.1687e - 8 1.4808e - 6 fexp 5.8117e - 8 1.2980e - 6 farcsech 5.3996e - 8 1.1011e - 6 acos[log] 5.3333e - 8 1.0726e - 6 acos[atan] 5.3114e - 8 1.0636e - 6 sin[acosh] 5.2941e - 8 1.0565e - 6 acos[asinh] 5.1480e - 8 9.9816e - 7 acos[exp] 5.1036e - 8 9.8109e - 7 acos[asech] 5.0695e - 8 9.6775e - 7 acos 5.0636e - 8 9.6618e - 7 acos[acos] 4.9480e - 8 9.2585e - 7 asech[exp] 4.9219e - 8 9.1771e - 7 asech[asech] 4.8700e - 8 9.0060e - 7 asech[acos] 4.8697e - 8 9.0126e - 7 Poly (39, -99, 85) 4.7582e - 8 8.8156e - 7 Power ([beta] = 0.29) 4.6192e - 8 8.2832e - 7 Pulse [P.sub.e] [E.sub.s] [s.sub.k] [alpha] = 0.25 [epsilon] = 0.2 t/T = 1 rcos 9.7462e - 4 0.0454 -0.9428 fsech 7.7201e - 4 0.0512 -0.9314 asech[log] 4.2640e - 4 0.0613 -0.8956 fexp 3.5678e - 4 0.0697 -0.8830 farcsech 2.8405e - 4 0.0738 -0.8664 acos[log] 2.7420e - 4 0.0751 -0.8630 acos[atan] 2.7117e - 4 0.0757 -0.8617 sin[acosh] 2.6878e - 4 0.0762 -0.8608 acos[asinh] 2.4946e - 4 0.0795 -0.8520 acos[exp] 2.4406e - 4 0.0807 -0.8489 acos[asech] 2.3971e - 4 0.0813 -0.8467 acos 2.3940e - 4 0.0817 -0.8460 acos[acos] 2.2747e - 4 0.0852 -0.8366 asech[exp] 2.2534e - 4 0.0862 -0.8340 asech[asech] 2.2053e - 4 0.0877 -0.8293 asech[acos] 2.2096e - 4 0.0880 -0.8289 Poly (39, -99, 85) 2.2060e - 4 0.0937 -0.8104 Power ([beta] = 0.29) 2.0300e - 4 0.0969 -0.8002 Pulse [a.sub.k] [alpha] = 0.25 t/T = 1 rcos 2.1095 fsech 2.1310 asech[log] 2.1938 fexp 2.2187 farcsech 2.2470 acos[log] 2.2531 acos[atan] 2.2555 sin[acosh] 2.2573 acos[asinh] 2.2731 acos[exp] 2.2787 acos[asech] 2.2826 acos 2.2838 acos[acos] 2.3008 asech[exp] 2.3054 asech[asech] 2.3138 asech[acos] 2.3145 Poly (39, -99, 85) 2.3468 Power ([beta] = 0.29) 2.3645 Table 6: ISI error probabilities of several Nyquist pulses for N = [2.sup.9] interfering symbols, [T.sub.f] = 40, M = 61, SNR = 15 dB, and [alpha] = 0.35. Pulse [P.sub.e] [P.sub.e] [alpha] = 0.35 [epsilon] = 0.05 [epsilon] = 0.1 cos 5.9997e - 8 1.3896e - 6 fsech 5.4002e - 8 1.0944e - 6 asech[log] 4.2145e - 8 6.2866e - 7 fexp 3.9253e - 8 5.4021e - 7 farcsech 3.5970e - 8 4.4580e - 7 acos[log] 3.5470e - 8 4.3365e - 7 acos[atan] 3.5310e - 8 4.3008e - 7 sin[acosh] 3.5182e - 8 4.2722e - 7 acos[asinh] 3.4124e - 8 4.041e - 7 acos[exp] 3.3806e - 8 3.9786e - 7 acos[asech] 3.3558e - 8 3.9255e - 7 acos 3.3527e - 8 3.9249e - 7 Poly (31, -80, 69) 3.2897e - 8 3.8388e - 7 Poly (53, -134,113) 3.2050e - 8 4.1318e - 7 Poly (36, -93, 80) 3.2461e - 8 3.7915e - 7 Poly (30.59, -78.27, 672) 3.2754e - 8 3.8186e - 7 acos[acos] 3.2753e - 8 3.7964e - 7 asech[exp] 3.2591e - 8 3.7775e - 7 asech[acos] 3.2264e - 8 3.7363e - 7 asech[asech] 3.2255e - 8 3.7275e - 7 Power ([beta] = 0.4) 3.1748e - 8 3.6256e - 7 Power ([beta] = 0.23) 3.0524e - 8 3.6990e - 7 Power ([beta] = 0.32) 3.0955e - 8 3.5466e - 7 Power ([beta] = 0.39) 3.1629e - 8 3.6109e - 7 Pulse [P.sub.e] [E.sub.s] [s.sub.k] [alpha] = 0.35 [epsilon] = 0.2 t/T = 1 cos 3.9084e - 4 0.0636 -0.8902 fsech 2.8000e - 4 0.0717 -0.8684 asech[log] 1.2567e - 4 0.0858 -0.8023 fexp 1.0129e - 4 0.0975 -0.7777 farcsech 7.6203e - 5 0.1033 -0.7475 acos[log] 7.3486e - 5 0.1051 -0.7411 acos[atan] 7.2778e - 5 0.1060 -0.7387 sin[acosh] 7.2196e - 5 0.1066 -0.7369 acos[asinh] 6.7653e - 5 0.1113 -0.7205 acos[exp] 6.6617e - 5 0.1123 -0.7147 acos[asech] 6.5582e - 5 0.1139 -0.7106 acos 6.5764e - 5 0.1144 -0.7094 Poly (31, -80, 69) 6.5629e - 5 0.1174 -0.6934 Poly (53, -134,113) 8.7720e - 5 0.1364 -0.6129 Poly (36, -93, 80) 6.6140e - 5 0.1203 -0.6801 Poly (30.59, -78.27, 672) 6.5544e - 5 0.1185 -0.6896 acos[acos] 6.4348e - 5 0.1193 -0.6918 asech[exp] 6.4445e - 5 0.1207 -0.6870 asech[acos] 6.4494e - 5 0.1232 -0.6774 asech[asech] 6.4110e - 5 0.1229 -0.6782 Power ([beta] = 0.4) 6.2244e - 5 0.1250 -0.6683 Power ([beta] = 0.23) 7.4044e - 5 0.1423 -0.5968 Power ([beta] = 0.32) 6.4326e - 5 0.1326 -0.6373 Power ([beta] = 0.39) 6.1940e - 5 0.1259 -0.6646 Pulse [a.sub.k] [alpha] = 0.35 t/T = 1 cos 2.2013 fsech 2.2397 asech[log] 2.3393 fexp 2.3876 farcsech 2.4301 acos[log] 2.4403 acos[atan] 2.4445 sin[acosh] 2.4476 acos[asinh] 2.4740 acos[exp] 2.4834 acos[asech] 2.4894 acos 2.4919 Poly (31, -80, 69) 2.5132 Poly (53, -134,113) 2.6351 Poly (36, -93, 80) 2.5326 Poly (30.59, -78.27, 672) 2.5196 acos[acos] 2.5199 asech[exp] 2.5277 asech[acos] 2.5427 asech[asech] 2.5412 Power ([beta] = 0.4) 2.5551 Power ([beta] = 0.23) 2.6616 Power ([beta] = 0.32) 2.6015 Power ([beta] = 0.39) 2.5606 Table 7: ISI error probabilities of several Nyquist pulses for N = [2.sup.9] interfering symbols, [T.sub.f] = 40, M = 61, SNR = 15 dB, and [alpha] = 0.5. Pulse [P.sub.e] [P.sub.e] [alpha] = 0.35 [epsilon] = 0.05 [epsilon] = 0.1 rcos 3.9723e - 8 5.4890e - 7 fsech 3.4949e - 8 4.1186e - 7 asech[log] 2.6157e - 8 2.1763e - 7 fexp 2.4134e - 8 1.8580e - 7 farcsech 2.1875e - 8 1.4916e - 7 acos[log] 2.1559e - 8 1.4514e - 7 acos[atan] 2.1462e - 8 1.4410e - 7 sin[acosh] 2.1386e - 8 1.4323e - 7 acos[asinh] 2.0758e - 8 1.3617e - 7 acos[exp] 2.0583e - 8 1.3446e - 7 Poly (25, -64, 55) 2.0574e - 8 1.3539e - 7 acos[asech] 2.0438e - 8 1.3273e - 7 acos 2.0431e - 8 1.3300e - 7 acos[acos] 2.0054e - 8 1.3014e - 7 asech[exp] 1.9992e - 8 1.3005e - 7 asech[acos] 1.9865e - 8 1.2958e - 7 asech[asech] 1.9845e - 8 1.2902e - 7 Power ([beta] = 0.37) 1.9451e - 8 1.2504e - 7 Power ([beta] = 0.32) 1.9356e - 8 1.2645e - 7 Power ([beta] = 0.37) 1.9451e - 8 1.2469e - 7 Pulse [P.sub.e] [E.sub.s] [s.sub.k] [alpha] = 0.35 [epsilon] = 0.2 t/T = 1 rcos 1.0217e - 4 0.0908 -0.7854 fsech 6.6009e - 5 0.1024 -0.7436 asech[log] 2.5364e - 5 0.1226 -0.6253 fexp 2.0878e - 5 0.1393 -0.5757 farcsech 1.5344e - 5 0.1476 -0.5231 acos[log] 1.4987e - 5 0.1502 -0.5114 acos[atan] 1.4953e - 5 0.1515 -0.5068 sin[acosh] 1.4911e - 5 0.1523 -0.5033 acos[asinh] 1.4609e - 5 0.1590 -0.4731 acos[exp] 1 .4657e - 5 0.1614 -0.4624 Poly (25, -64, 55) 1.5197e - 5 0.1615 -0.4575 acos[asech] 1.4563e - 5 0.1627 -0.4552 acos 1.4717e - 5 0.1635 -0.4527 acos[acos] 1.5328e - 5 0.1705 -0.4205 asech[exp] 1.5658e - 5 0.1724 -0.4116 asech[acos] 1 .6248e - 5 0.1761 -0.3942 asech[asech] 1 .6057e - 5 0.1756 -0.3957 Power ([beta] = 0.37) 1.6619e - 5 0.1825 -0.3586 Power ([beta] = 0.32) 1 .8540e - 5 0.1894 -0.3231 Power ([beta] = 0.37) 1.6619e - 5 0.1825 -0.3586 Pulse [a.sub.k] [alpha] = 0.35 t/T = 1 rcos 2.3562 fsech 2.4194 asech[log] 2.5316 fexp 2.6256 farcsech 2.6629 acos[log] 2.6768 acos[atan] 2.6836 sin[acosh] 2.6882 acos[asinh] 2.7239 acos[exp] 2.7368 Poly (25, -64, 55) 2.7316 acos[asech] 2.7434 acos 2.7479 acos[acos] 2.7850 asech[exp] 2.7958 asech[acos] 2.8119 asech[asech] 2.8148 Power ([beta] = 0.37) 2.8428 Power ([beta] = 0.32) 2.8759 Power ([beta] = 0.37) 2.8428 Table 8: Comparative performance of several Nyquist pulses for [alpha] = 0.25, [epsilon] = 0.1, and [epsilon] = 0.2. [alpha] = 0.25 Pulse [P.sub.e] [E.sub.s] [epsilon] = 0.1 asech[asech] 9.0060e - 7 0.0877 asech[acos] 9.0126e - 7 0.0880 [epsilon] = 0.2 asech[asech] 2.2053e - 4 0.0877 asech[acos] 2.2096e - 4 0.0880 [alpha] = 0.25 Pulse h ([epsilon]) [epsilon] = 0.1 asech[asech] 0.9819 asech[acos] 0.9819 [epsilon] = 0.2 asech[asech] 0.9289 asech[acos] 0.9288 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.1] [epsilon] = 0.1 asech[asech] 0.0711 asech[acos] 0.0711 [epsilon] = 0.2 asech[asech] 0.1182 asech[acos] 0.1182 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.2] [epsilon] = 0.1 asech[asech] 0.0873 asech[acos] 0.0871 [epsilon] = 0.2 asech[asech] 0.1432 asech[acos] 0.1430 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.3] [epsilon] = 0.1 asech[asech] 0.0920 asech[acos] 0.0920 [epsilon] = 0.2 asech[asech] 0.1545 asech[acos] 0.1545 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.4] [epsilon] = 0.1 asech[asech] 0.1034 asech[acos] 0.1035 [epsilon] = 0.2 asech[asech] 0.1764 asech[acos] 0.1766 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.5] [epsilon] = 0.1 asech[asech] 0.1133 asech[acos] 0.1135 [epsilon] = 0.2 asech[asech] 0.1944 asech[acos] 0.1949 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.7] [epsilon] = 0.1 asech[asech] 0.1186 asech[acos] 0.1189 [epsilon] = 0.2 asech[asech] 0.2043 asech[acos] 0.2049 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.10] [epsilon] = 0.1 asech[asech] 0.1287 asech[acos] 0.1289 [epsilon] = 0.2 asech[asech] 0.2231 asech[acos] 0.2235 Table 9: Comparative performance of several Nyquist pulses for [alpha] = 0.35, [epsilon] = 0.1, and [epsilon] = 0.2. [alpha] = 0.25 Pulse [P.sub.e] [E.sub.s] [epsilon] = 0.1 asech[asech] 3.7275e - 7 0.1229 asech[acos] 3.7363e - 7 0.1232 acos[asech] 3.9255e - 7 0.1139 acos 3.9249e - 7 0.1144 [epsilon] = 0.2 acos[acos] 6.4348e - 5 0.1193 asech[exp] 6.4445e - 5 0.1207 asech[asech] 6.4110e - 5 0.1229 asech[acos] 6.4494e - 5 0.1232 acos[asech] 6.5582e - 5 0.1139 acos 6.5764e - 5 0.1144 [alpha] = 0.25 Pulse h ([epsilon]) [epsilon] = 0.1 asech[asech] 0.9802 asech[acos] 0.9802 acos[asech] 0.9795 acos 0.9794 [epsilon] = 0.2 acos[acos] 0.9188 asech[exp] 0.9186 asech[asech] 0.9225 asech[acos] 0.9225 acos[asech] 0.9197 acos 0.9196 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.1] [epsilon] = 0.1 asech[asech] 0.0552 asech[acos] 0.0551 acos[asech] 0.0484 acos 0.0481 [epsilon] = 0.2 acos[acos] 0.0682 asech[exp] 0.0670 asech[asech] 0.0862 asech[acos] 0.0860 acos[asech] 0.0726 acos 0.0720 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.2] [epsilon] = 0.1 asech[asech] 0.0588 asech[acos] 0.0558 acos[asech] 0.0593 acos 0.0594 [epsilon] = 0.2 acos[acos] 0.0987 asech[exp] 0.0990 asech[asech] 0.0981 asech[acos] 0.0982 acos[asech] 0.0978 acos 0.0980 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.3] [epsilon] = 0.1 asech[asech] 0.0750 asech[acos] 0.0753 acos[asech] 0.0777 acos 0.0782 [epsilon] = 0.2 acos[acos] 0.1359 asech[exp] 0.1370 asech[asech] 0.1288 asech[acos] 0.1293 acos[asech] 0.1315 acos 0.1324 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.4] [epsilon] = 0.1 asech[asech] 0.0844 asech[acos] 0.0848 acos[asech] 0.0833 acos 0.0840 [epsilon] = 0.2 acos[acos] 0.1441 asech[exp] 0.1454 asech[asech] 0.1449 asech[acos] 0.1353 acos[asech] 0.1392 acos 0.1404 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.5] [epsilon] = 0.1 asech[asech] 0.0857 asech[acos] 0.0860 acos[asech] 0.0903 acos 0.0909 [epsilon] = 0.2 acos[acos] 0.1599 asech[exp] 0.1615 asech[asech] 0.1485 asech[acos] 0.1491 acos[asech] 0.1537 acos 0.1549 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.7] [epsilon] = 0.1 asech[asech] 0.0952 asech[acos] 0.0954 acos[asech] 0.1002 acos 0.1008 [epsilon] = 0.2 acos[acos] 0.1773 asech[exp] 0.1792 asech[asech] 0.1659 asech[acos] 0.1664 acos[asech] 0.1701 acos 0.1713 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.10] [epsilon] = 0.1 asech[asech] 0.1019 asech[acos] 0.1021 acos[asech] 0.1106 acos 0.1113 [epsilon] = 0.2 acos[acos] 0.2000 asech[exp] 0.2023 asech[asech] 0.1787 asech[acos] 0.1791 acos[asech] 0.1910 acos 0.1923 Table 10: Comparative performance of several Nyquist pulses for [alpha] = 0.5, [epsilon] = 0.1, and [epsilon] = 0.2. [alpha] = 0.25 Pulse [P.sub.e] [E.sub.s] [epsilon] = 0.1 acos[asech] 1.3273e - 7 0.1627 acos 1.3300e - 7 0.1635 [epsilon] = 0.2 acos[asinh] 1.4609e - 5 0.1590 acos[exp] 1.4657e - 5 0.1614 acos[asech] 1.4563e - 5 0.1627 acos 1.4717e - 5 0.1635 acos[acos] 1.5328e - 5 0.1705 asech[exp] 1.5658e - 5 0.1724 asech[asech] 1.6057e - 5 0.1756 asech[acos] 1.6248e - 5 0.1761 [alpha] = 0.25 Pulse h ([epsilon]) [epsilon] = 0.1 acos[asech] 0.9774 acos 0.9773 [epsilon] = 0.2 acos[asinh] 0.9126 acos[exp] 0.9121 acos[asech] 0.9118 acos 0.9117 acos[acos] 0.9102 asech[exp] 0.9098 asech[asech] 0.9091 asech[acos] 0.9090 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.1] [epsilon] = 0.1 acos[asech] 0.0324 acos 0.0321 [epsilon] = 0.2 acos[asinh] 0.0455 acos[exp] 0.0433 acos[asech] 0.0419 acos 0.0413 acos[acos] 0.0348 asech[exp] 0.0329 asech[asech] 0.0297 asech[acos] 0.0294 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.2] [epsilon] = 0.1 acos[asech] 0.0500 acos 0.0501 [epsilon] = 0.2 acos[asinh] 0.0772 acos[exp] 0.0769 acos[asech] 0.0763 acos 0.0765 acos[acos] 0.0752 asech[exp] 0.0749 asech[asech] 0.0738 asech[acos] 0.0740 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.3] [epsilon] = 0.1 acos[asech] 0.0571 acos 0.0576 [epsilon] = 0.2 acos[asinh] 0.0873 acos[exp] 0.0874 acos[asech] 0.0862 acos 0.0871 acos[acos] 0.0864 asech[exp] 0.0865 asech[asech] 0.0848 asech[acos] 0.0854 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.4] [epsilon] = 0.1 acos[asech] 0.0636 acos 0.0640 [epsilon] = 0.2 acos[asinh] 0.0993 acos[exp] 0.0997 acos[asech] 0.0994 acos 0.0999 acos[acos] 0.1008 asech[exp] 0.1012 asech[asech] 0.1008 asech[acos] 0.1013 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.5] [epsilon] = 0.1 acos[asech] 0.0670 acos 0.0673 [epsilon] = 0.2 acos[asinh] 0.1039 acos[exp] 0.1045 acos[asech] 0.1042 acos 0.1048 acos[acos] 0.1059 asech[exp] 0.1064 asech[asech] 0.1061 asech[acos] 0.1066 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.7] [epsilon] = 0.1 acos[asech] 0.0727 acos 0.0730 [epsilon] = 0.2 acos[asinh] 0.1136 acos[exp] 0.1145 acos[asech] 0.1145 acos 0.1151 acos[acos] 0.1174 asech[exp] 0.1182 asech[asech] 0.1187 asech[acos] 0.1191 [alpha] = 0.25 Pulse [d.sup.[epsilon].sub.10] [epsilon] = 0.1 acos[asech] 0.0782 acos 0.0784 [epsilon] = 0.2 acos[asinh] 0.1232 acos[exp] 0.1244 acos[asech] 0.1249 acos 0.1254 acos[acos] 0.1289 asech[exp] 0.1299 asech[asech] 0.1312 asech[acos] 0.1316 Table 11: ISI error probabilities of PS Nyquist pulses for N = [2.sup.9] interfering symbols, [T.sub.f] = 40, M = 61, SNR =15 dB, and [alpha] = 0.1. Pulse [P.sub.e] [P.sub.e] [alpha] = 0.1 [epsilon] = 0.05 [epsilon] = 0.05 PS (n = 0.05) 1.0098e - 7 4.6239e - 6 PS (n = 0.1) 1.0091e - 7 4.5877e - 7 PS (n = 0.2) 1.0159e - 7 4.6133e - 6 PS (n = 0.25) 1.0220e - 7 4.6579e - 6 PS (n = 0.3) 1.0291e - 7 4.7162e - 6 PS (n = 0.35) 1.0369e - 7 4.7847e - 6 PS (n = 0.4) 1.0452e - 7 4.8607e - 6 Poly (60, -152,130) 1.0355e - 7 4.8248e - 6 Power ([beta] = 0.07) 1.0073e - 7 4.5696e - 6 Pulse [P.sub.e] [E.sub.s] [alpha] = 0.1 [epsilon] = 0.05 PS (n = 0.05) 1.7038e - 3 0.0483 PS (n = 0.1) 1.6911e - 3 0.0468 PS (n = 0.2) 1.7031e - 3 0.0442 PS (n = 0.25) 1.7213e - 3 0.0430 PS (n = 0.3) 1.7448e - 3 0.0419 PS (n = 0.35) 1.7721e - 3 0.0408 PS (n = 0.4) 1.8023e - 3 0.0399 Poly (60, -152,130) 1.7843e - 3 0.0437 Power ([beta] = 0.07) 1.6840e - 3 0.0467 Table 12: ISI error probabilities of PS Nyquist pulses for N = [2.sup.9] interfering symbols, [T.sub.f] = 40, M = 61, SNR = 15 dB, and [alpha] = 0.25. Pulse [P.sub.e] [P.sub.e] [alpha] = 0.1 [epsilon] = 0.05 [epsilon] = 0.05 PS (n = 0.1) 4.6771e - 8 9.7161e - 7 PS (n = 0.2) 4.5937e - 8 8.8541e - 8 PS (n = 0.25) 4.5938e - 8 8.6765e - 7 PS (n = 0.3) 4.6119e - 8 8.6054e - 7 PS (n = 0.35) 4.6435e - 8 8.6133e - 7 PS (n = 0.4) 4.6852e - 8 8.6817e - 7 PS (n = 0.45) 4.7346e - 8 8.7972e - 7 PS (n = 0.5) 4.7897e - 8 8.9501e - 7 PS (n = 0.55) 4.8494e - 8 9.1334e - 7 Poly (39, -99, 85) 4.7582e - 8 8.8156e - 7 Power ([beta] = 0.29) 4.6192e - 8 8.2832e - 7 Pulse [P.sub.e] [E.sub.s] [alpha] = 0.1 [epsilon] = 0.05 PS (n = 0.1) 2.8226e - 4 0.1171 PS (n = 0.2) 2.3924e - 4 0.1104 PS (n = 0.25) 2.2871e - 4 0.1074 PS (n = 0.3) 2.2271e - 4 0.1047 PS (n = 0.35) 2.2006e - 4 0.1021 PS (n = 0.4) 2.1996e - 4 0.0997 PS (n = 0.45) 2.2187e - 4 0.0975 PS (n = 0.5) 2.2537e - 4 0.0953 PS (n = 0.55) 2.3018e - 4 0.0934 Poly (39, -99, 85) 2.2060e - 4 0.0937 Power ([beta] = 0.29) 2.0300e - 4 0.0969

Printer friendly Cite/link Email Feedback | |

Title Annotation: | Research Article |
---|---|

Author: | Alexandra, Nicolae Dumitru; Diaconu, Felix |

Publication: | Wireless Communications and Mobile Computing |

Article Type: | Report |

Date: | Jan 1, 2017 |

Words: | 9261 |

Previous Article: | Performance Analysis of MRC Receivers with Adaptive Modulation and Coding in Rayleigh Fading Correlated Channels with Imperfect CSIT. |

Next Article: | Trust on the Ratee: A Trust Management System for Social Internet of Vehicles. |

Topics: |