International Journal of Trend in Scientific

Research and Development (IJTSRD)

International Open Access Journal
ISSN No: 2456 - 6470 | www.ijtsrd.com | Volume - 2 | Issue – 3

Oscillation off Even Order Nonlinear Neutral Differential

Equations of E

G. Pushpalatha S. A. Vijaya Lakshmi

Assistant Professor, Department of Mathematics, Research Scholar, Department of Mathematics,
Vivekanandha College of Arts and Sciences for Vivekanandha College of Arts and Sciences
Science for
Women, Tiruchengode, Namakkal, Tamilnadu, India Women, Tiruchengode, Namakkal, Tamilnadu, India

ABSTRACT

This paper presently exhibits about the oscillation of (𝐸 ) 𝜌 ∈ 𝐶 1([𝑡 , ∞), 𝐸), 𝛾 ∈ 𝐶 ([𝑡 , ∞), 𝐸),
even order nonlinear neutral differential equations of
E of the form 𝛿 ∈ 𝐶([𝑡 , ∞), 𝐸), 𝜌′ (𝑡) ≥ 𝜌 > 0,

𝛾(𝑡) ≤ 𝑡, 𝛿(𝑡) ≤ 𝑡 , 𝜌 ∘ 𝛾 = 𝛾 ∘ 𝜌 ,
𝑒(𝑡)𝑧 ( )
(𝑡) + 𝑟(𝑡)𝑓(ℎ 𝛾(𝑡) + 𝑣
𝑣(𝑡)𝑓 𝛿(𝑡)
=0 𝜌 ∘ 𝛿 = 𝛿 ∘ 𝜌,
Where 𝑧(𝑡) = 𝑥(𝑡) + 𝑝(𝑡)𝑥 𝜌(𝑡) , 𝑛 ≥ 2, is a even
lim →∞ 𝛾(𝑡) = ∞ , lim (𝑡)
→∞ 𝛿( = ∞ , where 𝜌 is a
integer. The output we considered ∫ 𝑒 (𝑡)𝑑𝑡 = ∞, constant.
and ∫ 𝑒 (𝑡)𝑑𝑡 < ∞. This canon here extracted
(𝐸 ) 𝑓 ∈ 𝐶(𝐸, 𝐸) and
enhanced and developed a few known results in
literature.. Some model are given to embellish our 𝑓(𝑥)⁄𝑥 ≥ 𝑀 , 𝑀 > 0, for 𝑥 ≠ 0 , where 𝑀 , 𝑀 is
main results. constant.
INTRODUCTION Then the two cases are
We apprehensive with the oscillation theorems for the ∞
∫ 𝑑𝑡 = ∞ (2)
following half-linear even order neutral delay ( )

differential equation ∞
∫ 𝑑𝑡 < ∞, (3)
( )
′
𝑒(𝑡)𝑧 ( )
(𝑡) + 𝑟(𝑡)𝑓(ℎ 𝛾(𝑡) + 𝑣((𝑡)𝑓 𝛿(𝑡) =
By a solution 𝑧 of (1) a function be
0, 𝑡 ≥ 𝑡 , (1)
𝑒∈𝐶 ([𝑡 , ∞), 𝐸) for some 𝑡 ≥ 𝑡 ,
Where𝑧(𝑡) = 𝑥(𝑡) + 𝑝(𝑡)𝑥 𝜌(𝑡) , 𝑛 ≥ 2, is a even
integer .Every part of this paper, we assume that
that: Where 𝑧(𝑡) = 𝑥(𝑡) + 𝑎(𝑡)𝑥( ) 𝜌(𝑡) , has a property
1
𝑒𝑧 ∈ 𝐶 ([𝑡 , ∞), 𝐸) and satisfies (1) on (𝑡 , ∞).
(𝐸 ) 𝑒 ∈ 𝐶([𝑡 , ∞), 𝐸), 𝑒(𝑡) > 0, 𝑒′(𝑡) ≥ 0; Then (1) satisfies 𝑠𝑢𝑝{|𝑥)𝑡
{ 𝑡): 𝑡 ≥ 𝑇|} > 0 for all
𝑇 ≥ 𝑡 is called oscillatory.
(𝐸 ) 𝑝, 𝑞 ∈ 𝐶([𝑡 , ∞), 𝐸),
In certain case when 𝑛 = 2 the equation (1) lessen to
0 ≤ 𝑝(𝑡) ≤ 𝑝 < ∞, 𝑞(𝑡) > 0, where 𝑝 is a constant; the following equations

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 3 | Mar-Apr

Apr 2018 Page: 632
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
′ ′ If
(𝑒(𝑡) 𝑥(𝑡) + 𝑝(𝑡)𝑥 𝜌(𝑡) + 𝑟(𝑡)𝑓(ℎ 𝛾(𝑡) +
1
𝑣(𝑡)𝑓 𝛿(𝑡) = 0, 𝑡 ≥ 𝑡 lim 𝑖𝑛𝑓 𝑟(𝑠)𝑑𝑠 + lim 𝑖𝑛𝑓 𝑣(𝑠)𝑑𝑠 > ,
, →∞ ( ) →∞ ( ) 𝑒
(4) Then it has no finally positive solutions.

Where ∫ 𝑒 (𝑡)𝑑𝑡 = ∞, Main results

The main results which covenant that every solution
𝜌(𝑡) ≤ 𝑡 , 𝛾(𝑡) ≤ 𝑡, 𝛿(𝑡) ≤ 𝑡, of (1) is oscillatory
0 ≤ 𝑝(𝑡) ≤ 𝑝 < ∞.
(1) lim →∞ 𝑖𝑛𝑓 ∫ ( ) 𝐽(𝑠)𝑑𝑠 >
Then the oscillatory behavior of the solutions of the
neutral differential equations of the second order (2)lim 𝑠𝑢𝑝 𝐽(𝑠)𝑑𝑠 > 1
→∞ ( )
′ ′
(𝑒(𝑡) 𝑥(𝑡) + 𝑝(𝑡)𝑥 𝜌(𝑡) + 𝑟(𝑡)(ℎ 𝛾(𝑡) +
B.Theorem. 2.1
𝑣(𝑡) 𝛿(𝑡) = 0, 𝑡 ≥ 𝑡 (5)
∞
Assume that ∫ 𝑑𝑡 = ∞ holds. If
( )
Where ∫ 𝑒 (𝑠)𝑑𝑠 = ∞,
∞ ∞
0 ≤ 𝑝(𝑡) ≤ 𝑝 < ∞. 𝑆 (𝑡) + 𝑆 (𝑡) 𝑑𝑡 = ∞

The usual limitations on the coefficient of (5) be

𝜌(𝑡) ≤ 𝑡, 𝛾(𝑡) ≤ 𝜌(𝑡), 𝛿(𝑡) ≤ 𝜌(𝑡), Where 𝑆 (𝑡)= min {𝑟(𝑡), 𝑟(𝜌(𝑡))}
𝑆 (𝑡)= min{𝑣(𝑡), 𝑣(𝜌(𝑡))}, then every solutions of
𝛾(𝑡) ≤ 𝑡, 𝛿(𝑡) ≤ 𝑡, 0 ≤ 𝑝(𝑡) < 1, are not assumed. (1) is oscillatory.

𝜌 Could be a advanced argument and 𝛾, 𝛿 could be a Proof

delay argument ,
Suppose, on the contradictory,𝑥 is a nonoscillatory
Some known expand results are seen in [1,5]. Then solutions of (1). Without loss of generality, we may
the assume that there exists a constant 𝑡 ≥ 𝑡 , such that

Even-order nonlinear neutral functional differential 𝑥(𝑡) > 0, 𝑥(𝜌(𝑡)) > 0 and (𝛾(𝑡)) > 0 ,
equations
𝑥(𝛿(𝑡)) > 0 for all 𝑡 ≥ 𝑡 . Using the definitions of 𝑧
(n) and x is a eventually positive solution of (1). Then
(𝑥(𝑡) + 𝑝(𝑡)𝑥 𝜌(𝑡) )′ + 𝑟(𝑡)𝑓(ℎ 𝛾(𝑡) +
𝑣(𝑡)𝑓 𝛿(𝑡) = 0, 𝑡 ≥ 𝑡 (6) there exists 𝑡 ≥ 𝑡 , such that
,

Where n is even 0 ≤ 𝑝(𝑡) < 1 and 𝜌(𝑡) ≤ 𝑡. 𝑧(𝑡) > 0, 𝑧 ′ (𝑡) > 0, 𝑧 ( )
(𝑡) > 0 and 𝑧 (𝑡) ≤ 0 for
all 𝑡 ≥ 𝑡 .
(A) Lemma. 1
.Hencelim →∞ 𝑧(𝑡) ≠ 0.
The oscillatory behavior of solutions of the following
linear differential inequality Applying ( 𝐸 ) and (1) we get
′
𝑤 ′ (𝑡) + 𝑟(𝑡)ℎ(𝛾)) + 𝑣(𝑡)𝑔(𝛿(𝑡)) ≤ 0 𝑒(𝑡)𝑧 ( )
(𝑡) ≤ −𝑀 𝑟(𝑡)ℎ 𝛾(𝑡) < 0, 𝑡≥𝑡
Where 𝑟, 𝑣, 𝛾, 𝛿 ∈ 𝐶([𝑡 , ∞ )), ′
𝑒(𝑡)𝑧 ( )
(𝑡) ≤ −𝑀 𝑣(𝑡)𝑔 𝛿(𝑡) < 0, 𝑡 ≥ 𝑡 .
𝛾(𝑡)𝑡 , 𝛿(𝑡) ≤ 𝑡

lim 𝛾(𝑡) = ∞ , lim 𝛿(𝑡) = ∞ Therefore (𝑒(𝑡)𝑧 ( ) (𝑡)) is a nonincreasing

→∞ →∞ function. Besides, from the above inequality and the
definition of 𝑧, we get
Now integrating from 𝛾(𝑡) to 𝑡 and 𝛿(𝑡) 𝑡𝑜 𝑡

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 3 | Mar-Apr 2018 Page: 633
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
Case (1) ′
𝑒(𝑡)𝑧 ( )
(𝑡) + 𝑀 𝑉(𝑡)𝑧 𝛿(𝑡) +
′ ′
𝑒(𝑡)𝑧 ( )
(𝑡) + 𝑀 𝑟(𝑡)ℎ 𝛾(𝑡) + (𝑒 𝜌(𝑡) 𝑧 ( )
𝜌(𝑡) (10)
′
′( )
(𝑒 𝜌(𝑡) 𝑧 ( )
𝜌(𝑡) + Integrating (10) from 𝑡 𝑡𝑜 𝑡, we have
𝑀 𝑝 𝑟((𝜌(𝑡))ℎ(𝛾((𝜌(𝑡))) ≤ 0
𝑒(𝑠)𝑧 ( )
(𝑠))′𝑑𝑠
( )
(𝑒(𝑡)𝑧 (𝑡))′ + 𝑀 𝑅(𝑡)𝑧 𝛾(𝑡) +
(𝑒 𝜌(𝑡) 𝑧 ( ) (𝜌(𝑡)))′ ≤ 0 (7) +𝑀 𝑉(𝑠)𝑧 𝛿(𝑠) 𝑑𝑠

Integrating (7) from 𝑡 𝑡𝑜 𝑡, we have 𝑝 ′

+ (𝑒 𝜌(𝑠) 𝑧 ( )
𝜌(𝑠) 𝑑𝑠 ≤ 0
𝜌
(𝑒(𝑠)𝑧 ( )
(𝑠))′𝑑𝑠
Pointing that 𝜌′(𝑡) ≥ 𝜌 > 0

+𝑀 𝑅(𝑠)𝑧 𝛾(𝑠) 𝑑𝑠 𝑀 ∫ 𝑉(𝑠)𝑧 𝛿(𝑠) 𝑑𝑠 ≤ − ∫ 𝑒(𝑠)𝑧 ( )

(𝑠))′𝑑𝑠 −

𝑝 ∫ ′(
(𝑒 𝜌(𝑠) 𝑧 ( )
(𝜌(𝑠)))′ 𝑑𝑠 𝑑(𝜌(𝑠))
)
+ (𝑒 𝜌(𝑠) 𝑧 ( )
(𝜌(𝑠)))′ 𝑑𝑠
𝜌
≤
≤0
(𝑒(𝑡 )𝑧 ( ) 𝑡 −(𝑒(𝑡)𝑧 ( ) (𝑡)) +
Pointing that 𝜌′(𝑡) ≥ 𝜌 > 0 (𝑒 𝜌(𝑡 ) 𝑧 ( ) (𝜌(𝑡 ) −
𝑒 𝜌(𝑡) 𝑧 ( )
𝑒(𝑡) (11)
𝑀 𝑅(𝑠)𝑧 𝛾(𝑠) 𝑑𝑠 ≤ − 𝑒(𝑠)𝑧 ( )
(𝑠))′𝑑𝑠
Since 𝑧 ′ (𝑡) > 0 for 𝑡 ≥ 𝑡 . We can find a invariable
𝑝 1
− (𝑒 𝜌(𝑠) 𝑧 ( )
(𝜌(𝑠)))′ 𝑑𝑠 𝑑(𝜌(𝑠) 𝑐 > 0 such that
𝜌 ′
𝜌 (𝑠)
𝑧 𝛿(𝑡) ≥ 𝑐, 𝑡 ≥ 𝑡 .
≤
𝑒(𝑡 )𝑧 ( )
𝑡 −(𝑒(𝑡)𝑧 ( )
(𝑡)) + Then from (8) and the fact that (𝑒(𝑡)𝑧 ( ) (𝑡)) is
∞
(𝑒 𝜌(𝑡 ) 𝑧 ( )
(𝜌(𝑡 ) − nonincreasing, we obtain ∫ 𝑆 (𝑡) < ∞
(𝑒(𝜌(𝑡))𝑧 ( )
(𝑒(𝑡))) (8) (12)

Since 𝑧 ′ (𝑡) > 0 for 𝑡 ≥ 𝑡 . We can find a invariable We get inconsistency with (9),(12).
𝑐 > 0 such that ∞ ∞
𝑆 (𝑡) + 𝑆 (𝑡) = ∞
𝑧 𝛾(𝑡) ≥ 𝑐, 𝑡 ≥ 𝑡 .

Then from (8) and the fact that (𝑒(𝑡)𝑧 ( )

(𝑡)) is non Theorem. 2.2
increasing, we obtain ∞
Assume that ∫ 𝑑𝑡 = ∞ holds and 𝜌(𝑡) ≥ 𝑡. if
∞ ( )
∫ 𝑆 (𝑡) < ∞ (9) either
( ) ( )
lim 𝑖𝑛𝑓(∫ ( 𝑑𝑠 +
Case (2) →∞ ) ( ( ))
( ) ( ) ( )( )!
( ) ∫( 𝑑𝑠) > , (13)
(𝑒(𝑡)𝑧 (𝑡))′ + 𝑀 𝑣(𝑡)𝑔 𝛿(𝑡) ) ( ( ))
𝑝 ′
+ ′ (𝑒 𝜌(𝑡) 𝑧 ( ) 𝜌(𝑡) Or when 𝛾 , 𝛿 is increasing,
𝜌 (𝑡)
+ 𝑀 𝑝 𝑣((𝜌(𝑡))𝑔(𝛿((𝜌(𝑡))) ≤ 0

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 3 | Mar-Apr 2018 Page: 634
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
( ) ( )
lim →∞ 𝑠𝑢𝑝(∫ ( )
𝑑𝑠 + 𝑦(𝑡) ≤ 1 + 𝑥(𝑡) . (16)
( ( ))
( ) ( ) ( )( )!
∫( )
𝑑𝑠) > , (14) By combining (15) and (16), we get
( ( ))

Where (𝑡) = min 𝑀 𝑗(𝑡), 𝑀 𝑗 𝜌(𝑡) , ( ) ( )

𝑦 ′ (𝑡) + ( )! ( )
𝑦 𝛾(𝑡) +
then every solution of (1) is oscillatory . ( ) ( )
( )
𝑦 𝛿(𝑡) ≤0 (17)
Proof
Suppose, on the contrary 𝑥 is a oscillatory solution of Therefore , 𝑦 is a non negative solutions of (17).
(1). Without loss of generality, we may assume that
there exists a constant 𝑡 ≥ 𝑡 , such that t 𝑥(𝑡) > 0, Then there will be two cases

𝑥(𝜌(𝑡)) > 0 and 𝑥(𝛾(𝑡)) > 0 , 𝑥(𝛿(𝑡)) > 0 for all Case (1)
𝑡≥𝑡 . If
( ) ( )
Assume that 𝑥 ( ) (𝑡) is not consonantly zero on any lim →∞ 𝑖𝑛𝑓(∫ ( ) ( ( ))
𝑑𝑠 +
interval [𝑡 , ∞), and there exists a 𝑡 > 𝑡 . Such ( ) ( ) ( )( )!
( ) ( ) ∫( 𝑑𝑠) > holds then a
that𝑥 (𝑡)𝑢 (𝑡) ≤ 0 for all 𝑡≥𝑡 . If ) ( ( ))
lim →∞ 𝑥(𝑡) ≠ 0,then for every 𝜆, 0 < 𝜆 < 1, there constant be 0 < 𝜆 < 1, such that
exists𝑇 ≥ 𝑡 , such that for all 𝑡 ≥ 𝑇,
( ) ( )
lim →∞ 𝑖𝑛𝑓(∫ ( )( )!
( ( )
𝑑𝑠 +
( )
𝑥(𝑡) ≥ 𝑡 𝑢 (t) forever ( ) ( )
( )!
∫( ) ( )
𝑑𝑠)) > , (18)
0 < 𝜆 < 1 , we obtain
By lemma (1), (18) holds that (17) has negative
′
(𝑒(𝑡)𝑧 ( )
(𝑡))′ + (𝑒 𝜌(𝑡) 𝑧 ( )
𝜌(𝑡) + solutions, which is contradictory.

𝛾 (𝑡)𝐽(𝑡)𝑧 ( )
(𝛾((t)) + Case (2)
( )!
By the definition of 𝑦 and
( )!
𝛿 (𝑡)𝐾(𝑡)𝑧 ( )
(𝛾((t)) ≤ 0,
(
For every 𝑡 sufficiently large. 𝑒(𝑡)𝑧 ( ) (𝑡))′ + 𝑀 𝑅(𝑡)𝑧 𝛾(𝑡) +
(𝑒 𝜌(𝑡) 𝑧 ( ) (𝜌(𝑡)))′ ≤ 0
Let 𝑥(𝑡) = 𝑒(𝑡)𝑧 ( )
(𝑡) > 0. Then for all 𝑡 large
enough, we have we get ,
′ ′
𝑝 𝑦 ′ (𝑡) = 𝑥 ′ (𝑡) + 𝑥 𝜌(𝑡) ≤ −𝐽(𝑡)𝑧 𝛾(𝑡) −
𝑥(𝑡) + 𝑥 𝜌(𝑡)
𝜌 𝐾(𝑡)𝑧(𝛿(𝑡)) < 0 (19)
𝜆 𝛾 (𝑡)𝐽(𝑡) pointing that 𝛾(𝑡) ≤ 𝑡, 𝛿(𝑡) ≤ 𝑡, there exists 𝑡 ≥ 𝑡 ,
+ 𝑥 𝛾(𝑡) such that
(𝑛 − 1)! 𝑒(𝛾(𝑡))
𝜆 𝛿 (𝑡)𝐾(𝑡)
+ 𝑥 𝛿(𝑡) ≤ 0 𝑦 𝛾(𝑡) ≥ 𝑦(𝑡), 𝑦 𝛿(𝑡) ≥ 𝑦(𝑡), 𝑡 ≥ 𝑡 . (20)
(𝑛 − 1)! 𝑒(𝛿(𝑡))
(15) Integrating (17) from 𝛾(𝑡)𝑡𝑜 𝑡 and 𝛿(𝑡) to 𝑡 and
applying 𝛾 , 𝛿 is increasing, we have
Next , let us denote

𝑦(𝑡) = 𝑥(𝑡) + 𝑥 𝜌(𝑡) . Since 𝑥 is non increasing,

it follows from 𝜌(𝑡) ≥ 𝑡 that

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 3 | Mar-Apr 2018 Page: 635
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
𝑦(𝑡) − (𝑦 𝛾(𝑡) + 𝑦 𝛿(𝑡) ) + Or when 𝜌 𝜊𝛾 , 𝜌 𝜊𝛿 is increasing,
( ) ( )
∫
)! ( )
( 𝑦 𝛾(𝑠) + ( ) ( )
( ( ) lim →∞ 𝑠𝑢𝑝(∫ ( )
𝑑𝑠 +
( ) ( ) ( ( ))
∫( 𝑦 𝛿(𝑠) )𝑑𝑠 ≤ 0, 𝑡 ≥ 𝑡 . ( ) ( ) ( )( )!
) ( ) ∫ ( )
𝑑𝑠) > (24)
( ( ))
Thus
Where 𝐽, 𝐾 is defined as in theorem (2.2), then every
𝑦(𝑡) − (𝑦 𝛾(𝑡) + 𝑦(𝛿(𝑡))) + solution of (1) is oscillatory.
( ) ( )
((𝑦 𝛾(𝑡) ∫ ( )
+ Proof
( )! ( )
( ) ( )
𝑦 𝛿(𝑡) ) ∫ ( ) 𝑑𝑠 ≤ 0, 𝑡 ≥ 𝑡 Suppose, on the contrary 𝑥 is a oscillatory solution of
) ( )
(1). Without loss of generality, we may assume that
From the above the inequality,we get there exists a constant 𝑡 ≥ 𝑡 , such that

𝑦(𝑡) 𝑥(𝑡) > 0, 𝑥(𝜌(𝑡)) > 0 and

−1
(𝑦 𝛾(𝑡) + 𝑦(𝛿(𝑡))) (𝛾(𝑡)) > 0𝑥(𝛿(𝑡)) > 0 for all 𝑡 ≥ 𝑡 . Continuing as
𝜌 𝜆 𝛾 (𝑠)𝐽(𝑠) in the proof of the theorem (2.2), we have
+
𝑝 + 𝜌 (𝑛 − 1)! ( ) 𝑒 𝛾(𝑠) ′
𝛿 (𝑠)𝐾(𝑠) 𝑥(𝑡) + 𝑥 𝜌(𝑡) +
+ ) 𝑑𝑠 ≤ 0
( ) 𝑒 𝛿(𝑠)
( ) ( )
( )!
𝑥 𝛾(𝑡) +
( ( ))
From (20), we have ( ) ( )
( )!
𝑥 𝛿(𝑡) ≤ 0. Let
( ) ( ) ( ) ( ) ( ( ))
( )!
∫( ) ( )
+∫( ) ( )
) 𝑑𝑠 ≤
1,𝑡 ≥ 𝑡 (21) 𝑦(𝑡) = 𝑥(𝑡) + 𝑥 𝜌(𝑡) again. Since 𝑥 is non
increasing, it follows from𝜌(𝑡)) ≤ 0 that
Taking upper limits as 𝑡 → ∞ in (21) we get
( ) ( )
𝑦(𝑡) ≤ (1 + )𝑥 𝜌(𝑡) (25)
lim →∞ 𝑠𝑢𝑝(∫ ( )
𝑑𝑠 +
( ( ))
( ) ( ) ( )( )! By combining (15) and (24), we get
∫( )
𝑑𝑠) ≤ , (22)
( ( ))
( ) ( )
If (14) holds, we choose a constant 𝑦 ′ (𝑡) + ( )! ( )
𝑦 𝜌 𝛾(𝑡) +

0 < 𝜆 < 1 such that ( ) ( )

𝑦 𝜌 𝛿(𝑡) ≤0 (26)
( ( ))
( ) ( )
lim →∞ 𝑠𝑢𝑝(∫ ( )
𝑑𝑠 +
( ) Therefore, 𝑦 is a positive solution of (26). Now, we
( ) ( ) ( )( )!
∫( 𝑑𝑠) > , consider the following two cases , on (23) and (24)
) ( )
holds .
Which is in contrary with (22).
Case (1)
Theorem 2.3 If
∞ ( ) ( )
Assume that ∫ 𝑑𝑡 = ∞ holds and𝛾(𝑡) ≤ 𝜌(𝑡) ≤ lim →∞ 𝑖𝑛𝑓(∫ ( )
𝑑𝑠 +
( ) ( ( ))
( ) ( ) ( )( )!
𝑡, 𝛿(𝑡) ≤ 𝜌(𝑡) ≤ 𝑡. If either ∫( 𝑑𝑠) > holds then a
( ) ( ) ) ( ( ))
lim →∞ 𝑖𝑛𝑓(∫ ( )
𝑑𝑠 + constant be 0 < 𝜆 < 1, such that
( ( ))
( ) ( ) ( )( )!
∫ ( )
𝑑𝑠) > , (23)
( ( ))

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 3 | Mar-Apr 2018 Page: 636
 International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470

lim 𝑖𝑛𝑓(∫ ( (
( ) ( )
𝑑𝑠 + From (28) we get
→∞ )( )! ( )
𝛿𝑡𝑡𝛿𝑛−1𝑠𝐾𝑠𝑒𝛿𝑠𝑑𝑠))>1𝑒 (27) ( ) ( )
+
( )!
∫ ( ) ( ( ))
𝜌−1𝛿(𝑡)𝑡𝛿𝑛−1𝑠𝐾(𝑠)𝑒(𝛿𝑠)𝑑𝑠≤1, 𝑡≥𝑡2. (29)
Therefore (27) holds (26) has no positive solutions
which is a contradiction.
Taking the upper limit as 𝑡 → ∞ in (29), we get
Case (2)
( ) ( )
From (19) and the condition lim → 𝑠𝑢𝑝(∫ ( )
𝑑𝑠 +
( ( ))
𝛿(𝑡)𝑡𝛿𝑛−1𝑠𝐾(𝑠)𝑒(𝛿𝑠)𝑑𝑠)≤(𝑝0+𝜌0)(𝑛−1)!𝜆𝜌0,
𝛾(𝑡) ≤ 𝜌(𝑡), 𝛿(𝑡) ≤ 𝜌(𝑡), there exists
(30)
𝑡 ≥ 𝑡 , such that 𝑦 𝜌 𝛾(𝑡) ≥ 𝑦(𝑡) , Then the proof is similar to that of the theorem (2.2)
then it is contradiction to (24).
𝑦 𝜌 𝛿(𝑡) ≥ 𝑦(𝑡) 𝑡 ≥ 𝑡 . (28)
Reference
Integrating (26) from 𝜌 𝛾(𝑡) to𝑡, 𝜌 𝛿(𝑡) and 1) M.K.Grammatikopoulos, G. Ladas, A.
applying 𝜌 𝜊𝛾 is nondecreasing, then we get Meimaridou, Oscillation of second order neutral
delay differential equations, Rat. Mat. 1 (1985)
𝑦(𝑡) − 𝑦 𝜌 𝛾(𝑡) + 267-274.
( ) ( )
( )!
∫ ( ) ( )
𝑦 𝜌 𝛾(𝑡) + 2) J.Dzurina, I.P.Stavroulakis, Osillation criteria for
second order neutral delay differential equations,
𝜌−1𝛿(𝑡)𝑡𝛿𝑛−1𝑠𝐾(𝑠)𝑒(𝛿𝑠)𝑦𝜌−1𝛿𝑡𝑑𝑠≤ 0 , 𝑡≥𝑡2.
Appl.Math.Comput. 140 (2003) 445-453.

Thus 3) Z. L. Han, T. X. Li, S. R. Sun, Y.B. Sun, Remarks

on the paper [Appl. Math. Comput. 207(2009)
𝑦(𝑡) − 𝑦 𝜌 𝛾(𝑡) + 388-396], Appl. Math. Comput. 215 (2010) 3998-
( ) ( ) 4007.
( )!
𝑦 𝜌 𝛾(𝑡) ∫ ( )
+
( ( ))
4) Q. X. Zhang, J.R. Yan, L. Gao, Oscillation
𝑦𝜌−1𝛿𝑡𝜌−1𝛿(𝑡)𝑡𝛿𝑛−1𝑠𝐾(𝑠)𝑒(𝛿𝑠)𝑑𝑠≤0, 𝑡≥𝑡2. behavior of even – order nonlinear neutral delay
differential equations with variable coefficients,
From the inequality , we obtain Comput. Math.Appl. 59 (2010) 426-430.

𝑦(𝑡) 5) C. H.Zhang, T. X. Li, B. Sun, E. Thandapani, On

the oscillation of hogher- order half-linear delay
𝑦 𝜌 𝛾(𝑡) differential equations, Appl.Math. Lett. 24 (2011)
𝜌 𝜆 𝛾 (𝑠)𝐽(𝑠) 1618-1621.
−1
𝑝 + 𝜌 (𝑛 − 1)! ( ) 𝑒(𝛾(𝑠))
𝛿 (𝑠)𝐾(𝑠)
+ 𝑑𝑠 ≤ 0 .
( ) 𝑒(𝛿(𝑠))

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 3 | Mar-Apr 2018 Page: 637