Proceedings of the 19th World Congress

The International Federation of Automatic Control

Cape Town, South Africa. August 24-29, 2014

Comparison of Periodic and Event-Based

Sampling for Linear State Estimation ⋆
Bingchang Wang ∗,∗∗ , Minyue Fu ∗∗,∗∗∗
∗
School of Control Science and Engineering, Shandong University,
Jinan, 250061, China
∗∗
School of Electrical Engineering and Computer Science, University
of Newcastle, Callaghan, NSW, 2308, Australia
∗∗∗
Department of Control Science and Engineering, Zhejiang
University, Hangzhou, 310058, China

Abstract: In this paper, the state estimation problem for continuous-time linear systems with
two types of sampling is considered. First, the optimal state estimator under periodic sampling
is presented. Then the state estimator with event-based updates is designed, i.e., when an event
occurs the estimator is updated linearly by using the measurement of output, while between the
consecutive event times the estimator is updated by minimum mean-squared error criteria. The
average estimation errors under both sampling schemes are compared quantitatively for first and
second order systems, respectively. A numerical example is given to compare the effectiveness
of two state estimators.

Keywords: Event-based estimation, Kalman filter, periodic sampling, event-based sampling,

average estimation errors

1. INTRODUCTION event-based sampling for state estimation of higher order

Recently, event-based control and estimation have received
a lot of attention in the system and control community
(Lemmon, 2010; Heemels et al., 2012). Event-based mecha-
nism occurs naturally in many situations, such as relay sys-
tems (Dodds, 1981), pulse modulation (Holmes and Lipo,
2003), and biological systems (Keener and Sneyd, 2008);
particularly it has been boosted in wireless sensor networks
where the communication resource is limited and even
scare (Xu and Hespanha, 2004; Imer and Basar, 2005).
Åström and Bernhardsson (2002) highlighted some advan-
tages of event-based sampling over periodic sampling and
motivated the development of systematic design and anal-
ysis of event-based controllers and estimators (Tabuada,
2007; Henningsson et al., 2008; Rabi et al., 2008; Li et al.,
2010; Wang and Lemmon, 2011; Meng and Chen, 2012;
Meng et al., 2013; Wang et al., 2014). In this paper, we
focus on the event-based estimation problem.
To reduce data transmission rate in sensor networks,
some estimation schemes and algorithms on event-based
mechanism such as “Send on Delta” were presented and
considered (Miskowicz, 2006; Suh et al., 2007). Åström
(2007) provided some comments on Kalman filter under
event-based sampling. Rabi (2006) and Rabi et al. (2012)
investigated the state estimation problem for first order
linear systems with a constraint on the sampling rate,
and compared the estimation errors under event-based and
periodic sampling. However, to our knowledge, there has
been no work carrying out the comparison of periodic and
⋆ This work was supported by the Fundamental Research Funds of
Shandong University under Grant 2014TB007.
Email: bcwang@sdu.edu.cn, minyue.fu@newcastle.edu.au.
systems, which is definitely worth studying.
The previous work (Sijs and Lazar, 2011; Wu et al., 2013;
You and Xie, 2013) gave some approximating algorithms
for event-based estimator under the assumption that the
predicted estimate or the innovation is Gaussian. Sijs and
Lazar (2011) considered the state estimation under general
event based sampling, and provided a stable approxima-
tion algorithm by using Gaussian sums. Wu et al. (2013),
and You and Xie (2013) devised scheduling schemes that
the output is communicated based on measurement inno-
vation, and provided recursive algorithms for the discrete-
time state estimation with the assumption of Gaussian
prediction. However, the above Gaussian assumption is
not guaranteed to be true. Without the assumption, Shi
et al. (2013) considered the discrete-time event-triggered
state estimation problem in the framework of maximum
likelihood estimation. In this paper, some results for
continuous-time linear state estimation are provided with-
out the Gaussion assumption.
The objective of the paper is to compare periodic
and event-based sampling for the state estimation of
continuous-time linear stochastic systems. First, we present
the optimal state estimator under periodic sampling. Then
we design the state estimator with event-based updates.
Specifically, when an event occurs the estimator is updated
linearly by using the upcoming measurement, while be-
tween the consecutive event times the estimator is updated
by MMSE (minimum mean-squared error) criteria. The
main idea of the event-based sampling is that the output
measurement is transmitted to the estimator only if the
measurement innovation is significant enough. Later, we
compare the mean-square estimate errors of two state

978-3-902823-62-5/2014 © IFAC 5508

19th IFAC World Congress
Cape Town, South Africa. August 24-29, 2014

∫ τk+1
estimators under the same average sampling rates for
first order systems and symmetric second order systems, x(τk+1 ) = eA(τk+1 −τk ) x(τk ) + eA(τk+1 −s) DdW (s),
τk
respectively. Finally, we give a numerical example to com- y(τk ) = Cx(τk ), k = 0, 1, . . . .
pare the estimation errors of two state estimators. (5)
The following notations will be used in the paper. For a First, consider the case of periodic sampling, i.e., τk = kh.
sequence of numbers aj , j = 1, . . . , m, diag{a1 , . . . , am } Following the argument to the standard Kalman filter
denotes the diagonal matrix with aj in the diagonal derivation (Anderson and Moore, 1979; Åström, 2006), we
and zero elsewhere. For a stochastic process x(t), t ≥ 0, get the algorithm for the periodic estimator:
σ(x(s), s ≤ t) denotes the σ-algebra generated by x(s), s ≤
t. E[·] and E[·|·] denote the mathematical expectation and Algorithm I. Periodic estimator:
the conditional expectation, respectively.
x̂[(k + 1)h] = x̂′ [(k + 1)h]
{ }
2. PROBLEM FORMULATION + Kk+1 y[(k + 1)h] − C x̂′ [(k + 1)h] ,
x̂′ [(k + 1)h] = eAh x̂(kh),
′ ′
Consider the following linear stochastic system: Kk+1 = Pk+1 C T (CPk+1 C T )−1 ,
′ ′
Pk+1 = Pk+1 − Kk+1 CPk+1 ;
dx(t) = Ax(t) + DdW (t), (1) ′ T
Pk+1 = eAh Pk eA h
y(t) = Cx(t), (2) ∫ (k+1)h
T

where x(t) ∈ R , y(t) ∈ R are the state and the

n m
output, respectively. E[x(0)] = µ, E[(x(0) − µ)(x(0) −
µ)T ] = Σ > 0. The stochastic disturbance W (t) ∈ Rd
is a standard Wiener process defined on a probability
space (Ω, F, P). Assume C ∈ Rm×n has full row rank.
(A, D) is controllable. The output y(t) is transmitted
to the estimator by a sensor with limited computational
capability at sampling times τ0 , . . . , τN , . . ..
In this paper, we will design and compare two types of
estimators for the state x(t) based on the measurements
y(τ0 ), . . . , y(τk ), and sampling times τ0 , . . . , τk , where τk ≤
t < τk+1 , k = 0, 1, . . .. To be specific, for the first estimator
the sampling times are deterministic and periodic, while
for the second estimator the sampling times are a sequence
of random times based on the event occurrence.
3. STATE ESTIMATORS
From (1), we have for any k = 0, 1, . . .,
∫ t
x(t) = eA(t−τk ) x(τk ) + eA(t−s) DdW (s), τk ≤ t. (3)
τk
Let Ft = σ{x(s), s ≤ t}, and Yt = σ{y(s), s ≤ t}. Let
Ht = σ{y(τ0 ), . . . , y(τk ), τ0 , . . . , τk } for τk ≤ t < τk+1 .
Noting Ht ⊂ Yt ⊂ Ft and Ht = Hτk , τk ≤ t < τk+1 , from
+ eA[(k+1)h−s] DDT eA [(k+1)h−s] ds,
kh
x̂′ (0) = µ, P0′ = Σ. (6)
We now consider the case that the sampling times are the
following event-based times:
τ0 = 0,
τk+1 = inf{t > τk ; ∥z(t−)∥ ≥ δ}, k = 0, 1, . . . , (7)
where δ > 0 is a tuning parameter, z(t) = y(t) − C x̂(t) is
the so-called innovation process, and z(t−) = lims↑t z(s)
is the left limit of z(t).
Remark 3.1. The main idea of the specification for the
event-based times τk , k = 0, 1, . . . is that the output is
transmitted to the estimator only when the measurement
innovation (the error between the measurement and the
predicted value of output) is large enough; otherwise it is
not regarded “significant enough” and hence not be sent
to the estimator. It is still not clear whether the innovation
process z(t) is approximately Gaussian or not. However,
∥z(τk −)∥ = δ, and z(τk −) is non-Gaussian. Actually, for
the one-dimensional case, z(τk −) is binomial, and for the
two-dimensional case, z(τk −) is supported on a circle.
From (4), we have
x̂(τk+1 −) = lim x̂(t) = eA(τk+1 −τk ) x̂(τk ).
∆
(8)
t↑τk+1

(3) we obtain that At the time τk+1 , an event occurs and a new measurement
is received by the estimator. Then by Theorem 3.2.1 in
Anderson and Moore (1979), we get the linear MMSE
E[x(t)|Ht ] = E[eA(t−τk ) x(τk )|Hτk ]
∫ t update as follows:
[  ]
+E E[ eA(t−s) DdW (s)|Fτk ]Hτk x̂(τk+1 ) = x̂(τk+1 −) + Kk+1 z(τk+1 −), k = 0, 1, . . . (9)
τk where
= eA(t−τk ) E[x(τk )|Hτk ], τk ≤ t < τk+1 . (4) −
Kk+1 = Pk+1 −
C T (CPk+1 C T )−1 , (10)
From the fundamental result of state estimation (Åström,
−
= eA(τk+1 −τk ) Pk eA (τk+1 −τk )
T
2006; Oksendal, 2003), we have the MMSE (minimum Pk+1
mean-squared error) update between two consecutive sam- ∫ τk+1
eA(τk+1 −s) DDT eA (τk+1 −s) ds,
T
pling times: + (11)
τk
x̂(t) = e x̂(τk ), τk ≤ t < τk+1 ,
A(t−τk )
Pk = Pk− − Kk CPk− . (12)
where x̂(t) is the estimator for x(t). Here, (11) is obtained from
∫ τk+1
In what follows, we develop the recursive algorithms for x̃(τk+1 −) = eA(τk+1 −τk )
x̃(τk ) + eA(τk+1 −s) DdW (s),
x̂(τk ), k = 0, 1, . . . . By (2) and (3), it follows that τk

5509
19th IFAC World Congress
Cape Town, South Africa. August 24-29, 2014

n−1 ∫ [ ∫ t 2 ]
which follows by (5) and (8). 1 ∑ (k+1)h  
= lim sup E  σea(t−s) dw(s) dt.
By the above result we get the following algorithm for the n→∞ nh kh kh
k=0
event-based estimator.
Thus, we have for the case a = 0,
Algorithm II. Event-based estimator: n−1 ∫
1 ∑ (k+1)h 2 σ2 h
[ ] JP = lim sup σ (t − kh)dt = ,
x̂(τk ) = x̂(τk −) + Kk y(τk ) − C x̂(τk −) , n→∞ nh kh 2
k=0

x̂(τk+1 −) = eA(τk+1 −τk ) x̂(τk ), and for the case a ̸= 0,

Kk = Pk− C T (CPk− C T )−1 , ∫
1 ∑ 2 (k+1)h e2a(t−kh) − 1
n−1
Pk = Pk− − Kk CPk− , JP = lim sup σ dt
n→∞ nh kh 2a
−
= eA(τk+1 −τk ) Pk− eA (τk+1 −τk )
T
k=0
Pk+1
∫ τk+1 σ 2 (e2ah − 2ah − 1)
= .
eA[τk+1 −s] DDT eA [τk+1 −s] ds,
T
+ 4a2 h
τk
x̂(0−) = µ, P0− = Σ. (13) In what follows, we calculate the average estimation error
under the event-based sampling. In this case, the sampling
Different from Algorithm I, the updating times τk , k = times are τ0 = 0, and
1, 2, . . . in Algorithm II are random. From (7), they depend τk+1 = inf{t > τk ; |z(t−)| ≥ δ}, k = 0, 1, . . . ,
on the output y(t), which implies that the gains Kk , k = ∫t
where z(t) = y(t) − cx̂(t) = σ τk ea(t−s) dw(s), τk ≤ t <
1, 2, . . . are also random and dependent on y(t).
τk+1 . It can be verified that z(τk ) = 0, and z(t) satisfies
Remark 3.2. Noting τ0 = 0, Σ > 0, it follows that
z(τ0 ) = y(τ0 ) − C x̂(τ0 ) = 0, which leads to τ1 > dz(t) = az(t) + cσdw(t), τk ≤ t < τk+1 . (17)
0. Since (A, D) is controllable, it follows that for any Let Zt = σ(z(s), s ≤ t). Noting z(t) is a time-homogeneous
∫t T strong Markovian process (Oksendal, 2003) and z(τk ) = 0,
t > 0, 0 eAs DDT eA s ds is positive definite, which gives
∫ τ1 A(τ −s) k = 0, 1, 2, . . ., we obtain that for any Borel-measurable
that τ0 e 1 DDT eA (τ1 −s) ds is positive definite. This
T
function f ,
together with the fact that C has full row rank implies that [ ∫ τk+1  ] [ ∫ τk+1  ]
 
CP1− C T is invertible and positive definite. From (13), it E f (z(t))dtZτk = E f (z(t))dtz(τk )
follows that z(τ1 ) = y(τ1 ) − C x̂(τ1 ) = 0, which together τk
[ ∫ τ1
τk
 ]
with (7) leads to τ2 > τ1 . Thus, by the induction we obtain 
=E f (z(t))dtz(τ0 )
that τk+1 is strictly greater than τk , which guarantees the
[ ∫ τ1
τ0
sequence of sampling times τk , k = 0, 1, . . . are well defined. ]
=E f (z(t))dt .
4. THE CASE OF FIRST ORDER SYSTEMS ∫τ 0
Thus, τkk+1 f (z(t))dt, k = 0, 1, . . . are independent and
In this section, consider the first order system described identically distributed. From this together with (16), the
by average estimation error is
∫
1 ∑ [ τk+1 ]
n−1
dx(t) = ax(t)dt + σdw(t), (14)
JE = lim sup 2
E |z(t)|2 dt
y(t) = cx(t), (15) n→∞ c τn τk
k=0
∑ [ ∫ τk+1 ]
where x(t), y(t) ∈ R, c ̸= 0, and σ ̸= 0. w(t) is a one- n 1
n−1

dimensional Wiener process. We will compare the average = lim sup 2 E |z(t)|2 dt
n→∞ c τn n τk
estimation errors for both sampling schemes. [∫τ ]
k=0
E 0 1 |z(t)|2 dt
For the system above, by Algorithm II we obtain that = . (18)
1 1 c2 E[τ1 ]
x̂(τk ) = y(τk ), Kk = , Pk = 0. From the probabilistic representation of solutions to dif-
c c
ferential equations (See Lemma 3.1 in Wang et al. (2014)
From (4), it follows that
or Oksendal (2003) ), it follows that E[τ1 |z(0) = x] is the
1 solution of
x̂(t) = ea(t−τk ) y(τk ), τk ≤ t < τk+1 .
c c2 σ 2 ′′
h (x) + axh′ (x) = −1,
We now check the average estimation errors 2
[∫ ] with boundary conditions h(δ) = h(−δ) = 0, which gives
T
1 ∫ δ ∫ y
J = lim sup E |x(t) − x̂(t)| dt ,
2
(16) 1 [ a(y 2 − z 2 ) ]
T →∞ T 0 E[τ1 ] =2 2 2
exp − dzdy
0 c σ 0 c2 σ 2
for periodic and event-based sampling, respectively. First, ∑∞
under the periodic sampling scheme (i.e., τk = kh), the 22k−1 (−a)k−1 (k − 1)!δ 2k
= .
average estimation error is (2k)!
k=1

n−1 ∫
Similarly, we have
1 ∑ (k+1)h [ ] [ ∫ τ1 ] ∫ δ ∫ y
z2 [ a(y 2 − z 2 ) ]
JP = lim sup E |x(t) − x̂(t)|2 dt E |z(t)| dt = 2
2
exp − dzdy.
n→∞ nh kh c2 σ 2 c2 σ 2
k=0 0 0 0

5510
19th IFAC World Congress
Cape Town, South Africa. August 24-29, 2014

n−1 ∫
5 1 ∑ (k+1)h [ ]
JP = lim sup E ∥x(t) − x̂(t)∥2 dt
4.5
n→∞ nh
a=1 k=0 kh
∫ ∫ 2 ]
1 ∑ (k+1)h [ t a(t−s)
4
n−1

3.5
a=0
= lim sup 2E  e dw(s) dt
n→∞ nh
k=0 kh kh
JP/JE

{
3

2.5 h a=0
= e2ah − 2ah − 1
a ̸= 0.
2 a= −1

1.5 2a2 h
1
0 0.5
h or Eτ1
1 1.5 In what follows, we calculate the average estimation error
under the event-based sampling. In this case, the sampling
times are τk+1 = inf{t > τk ; ∥z(t−)∥ ≥ δ}, where z(t) =
Fig. 1. JP /JE for a = −1, 0, 1 with the same average
x(t) − x̂(t). Then
sampling period
(∫ t ∫ t )T
From this together (18), we get z(t) = ea(t−s) dw1 (s), ea(t−s) dw2 (s)
∫δ∫y 2 [ a(y 2 −z 2 ) ]
τk τk

0 0
z exp − c2 σ 2 dzdy satisfies
JE = ∫ δ ∫ y [ a(y 2 −z 2 ) ]
,
c2 0 0 exp − dzdy dz(t) = Az(t) + dw(t), z(τk ) = (0, 0)T , τk ≤ t < τk+1 .
c2 σ 2
which coincides with the result in (Rabi, 2006, Chapter 3). As in Gardiner (2004) and Meng and Chen (2012), set
Particularly, for the case a = 0, z1 (t) = r(t)cos[ϕ(t)], z2 (t) = r(t)sin[ϕ(t)],
δ2 δ2 with √
E[τ1 ] = 2 2 , JE = 2 .
c σ 6c r(t) = z12 (t) + z22 (t).
We now compare the estimation effectiveness under both Then by using Ito’s formula, we have that
sampling schemes. Specifically, the estimation error with 1
dr(t) = [ar(t)+ ]dt + dv(t), r(τk ) = 0, τk ≤ t < τk+1 ,
periodic sampling is compared to one with event-based 2r(t)
sampling under the assumption that average sampling where
rates are the same, i.e. h = hE . For the case a = 0, we
have h = E[τ1 ] = δ 2 /(c2 σ 2 ) and v(t) = w1 (t)cos[ϕ(t)] + w2 (t)sin[ϕ(t)].
It can be verified that v(t) is a standard Wiener process,
JP δ 2 /2c2
= 2 2 = 3. since it is an orthogonal transformation of w(t).
JE δ /6c
From Lemma 3.1 in Wang et al. (2014) or Oksendal (2003),
The ratio JP /JE as a function of h or E[τ1 ] is plotted in
we obtain that E[τ1 |r(0) = x] satisfies
Figure 1 for a = −1, 0, 1. The figure shows that event-
based sampling gives substantially smaller estimation er- 1 ′′ 1 ′
g (x) + (ax + )g (x) = −1,
rors under the same sampling rates. It can be seen that 2 2x
the ratio for unstable systems is larger than one for stable with boundary conditions g(δ) = g(−δ) = 0, which gives
systems.
∫ ∫
δ
z y[ ]
E[τ1 ] = 2 exp − a(y 2 − z 2 ) dzdy
0 0 y
∞
∑ (−a)k−1 δ 2k
= .
5. THE CASE OF SYMMETRIC SECOND ORDER 2kk!
k=1
SYSTEMS
Similarly, we have
In this section, consider the following second order system [∫ τ1 ] ∫ δ
z3
∫
[y ]
described by E ∥z(t)∥2 dt = 2 exp −a(y 2 − z 2 ) dzdy
0 0 0 y
∞
∑
dx(t) = Ax(t)dt + Ddw(t), (−a)k−1 δ (2k+2)
= .
y(t) = Cx(t), (19) (2k + 2)(k + 1)!
k=1
where x(t), y(t) ∈ R , A = diag{a, a}, C = diag{c, c}
2
From this together (18), we get
and D = diag{σ, σ}. w(t) = (w1 (t), w2 (t))T is a two- ∫ δ ∫ y z3 [ ]
exp −a(y 2 − z 2 ) dzdy
dimensional Wiener process. Without loss of generality, 0 0 y
JE = ∫ δ ∫ y [ ] .
assume c = σ = 1. In this case, by Algorithm II and (4),
0 0 y
z
exp −a(y 2 − z 2 ) dzdy
it follows that
Particularly, for the case a = 0,
Kk = I, Pk = 0, x̂(τk ) = x(τk ),
δ2 δ2
x̂(t) = eA(t−τk ) x(τk ), τk ≤ t < τk+1 . E[τ1 ] = , JE = .
2 4
First, for the case of the periodic sampling the average We now compare the estimation errors for both sampling
estimation error is under the assumption that average sampling rates are

5511
19th IFAC World Congress
Cape Town, South Africa. August 24-29, 2014

∫ t
3

2.8
= (t − τk )x̃2 (τk ) + (t − s)dw2 (s)
τk
2.6 ∫ t
2.4
= [x̃2 (τk ) + w2 (s)]ds, (21)
2.2 a=1
τk
JP/JE

where x̃(τk ) = x(τk ) − x̂(τk ). From this together with (7),

2
a=0
{ ∫ t  }
1.8

1.6 a= −1  
1.4
τk+1 = inf t > τk ;  [x̃2 (τk ) + w2 (s)]ds = δ .
τk
1.2

1
0 0.2 0.4
h or Eτ1
0.6 0.8 1 Take δ = 0.3. By implementing Algorithm II, we get the
trajectory of the event-based estimator, which together
Fig. 2. JP /JE for a = −1, 0, 1 with the same average with the state x(t) is shown in Figure 3. Meanwhile, we
sampling period have limk→∞ τk /k = 0.64. Let h = 0.64. From Algorithm I,
we get the trajectory of the the periodic estimator, which is
the same, i.e. h = E[τ1 ]. For the case a = 0, we have also plotted in Figure 3. It can be seen that the event-based
h = E[τ1 ] = δ 2 /2, and estimator is more consistent with the state trajectory than
JP δ 2 /2 the periodic estimator, irrespective of the first state or the
= 2 = 2. second state.
JE δ /4
The ratio JP /JE as a function of h or E[τ1 ] is plotted in 6
the state
Figure 2 for a = −1, 0, 1. The figure shows that event- 5 event based estimator
periodic estimator
based sampling gives smaller estimation errors and the 4
ratio for unstable systems is larger than one for stable
3
systems. Different from the case of first order systems,
the improvement of event-based sampling over periodic 2

sampling declines for unstable systems when the average 1

sampling period exceeds around 0.7. 0

−1

−2
0 2 4 6 8 10
t/s

(a) The first state

3
the state
6. AN ILLUSTRATIVE EXAMPLE event based estimator
2 periodic estimator

In this section, an example of the double integrator is 1

provided to demonstrate and compare the effectiveness of

0
two state estimators.
−1
For the system in (1)-(2), set
( ) ( )
01 0 0 −2

A= , D= , and C = (1 0).
00 0 1 −3
0 1 2 3 4 5 6 7 8 9 10
t/s
Then it follows that
( ) ( ) (b) The second state
0 1 0
dx(t) = x(t)dt +
0 0 dw2 (t) Fig. 3. Trajectories of the state, event-based estimator and
y(t) = (1 0) x(t). (20) periodic estimator
Here x(t) = (x1 (t), x2 (t))T is characterized as the position The squared estimation errors ∥x̃(t)∥2 under event-based
and speed of an object, while only the position can be and periodic sampling are shown in Figure 4. It can be
measured. Noting seen that the average estimation errors under event-based
( )
1 t sampling are smaller compared with the periodic sampling.
At
e = , C(x(τk ) − x̂(τk )) = 0,
01 Indeed, the time-average value of squared estimation errors
we get that for τk ≤ t < τk+1 , k = 0, 1, . . ., under event-based and periodic samplings are 0.4107 and
0.5636, respectively.
∫ t
z(t) = CeA(t−τk ) (x(τk ) − x̂(τk )) + C eA(t−s) Ddw(s) 7. CONCLUDING REMARKS
( )( ) τk
1 t − τk x̃1 (τk )
= (1 0) This paper considered the state estimation problem of
0 1 x̃2 (τk )
∫ t( )( ) continuous-time linear systems with two types of sampling.
1 t−s 0 We first presented the optimal state estimator under
+(1 0)
τk
0 1 dw2 (s) periodic sampling, and designed the state estimator with

5512
19th IFAC World Congress
Cape Town, South Africa. August 24-29, 2014
7
periodic
event based
6
5 trol Conference, 1029–1034.
squared estimation errors
4
3
2 estimation, and optimization. In Networked Control
1
0
0 2 4
t/s
6 8
Fig. 4. Squared estimation errors under event-based and
periodic sampling
event-based updates. Then, we compared the average
estimation errors for both sampling schemes in first and
second order systems, respectively. A numerical example
was provided to compare the effectiveness of two state
estimators. It can be shown that in these cases the event-
based sampling does outperform periodic sampling for
state estimation. Specifically, for critically stable first order
systems, the ratio of the average estimation error with
periodic sampling to one with event-based sampling is 3;
for critically stable symmetric second order systems, the
ratio is 2. However, due to severe mathematical difficulties,
the quantitative comparison of average estimation errors
for the general case under both sampling schemes was not
provided in this paper, which needs to be investigated
further.
ACKNOWLEDGEMENTS
We would like to thank Professor Tongwen Chen for his
many insightful discussions and suggestions.
REFERENCES
Anderson, B. and Moore, J. (1979). Optimal Filtering.
Prentice-hall Englewood Cliffs, NJ.
Åström, K. (2006). Introduction to Stochastic Control
Theory. Courier Dover Publications.
Åström, K. and Bernhardsson, B. (2002). Comparison of
Riemann and Lebesgue sampling for first order stochas-
tic systems. In Proceedings of the 41st IEEE Conference
on Decision and Control, volume 2, 2011–2016.
Åström, K. (2007). Event based control. Analysis and De-
sign of Nonlinear Control Systems: In Honor of Alberto
Isidori, 127–147.
Dodds, S. (1981). Adaptive, high precision, satellite atti-
tude control for microprocessor implementation. Auto-
matica, 17(4), 563–573.
Gardiner, C. (2004). Handbook of Stochastic Methods: For
Physics, Chemistry and the Natural Sciences.
Heemels, W., Johansson, K.H., and Tabuada, P. (2012).
An introduction to event-triggered and self-triggered
control. In Proceedings of the 51th IEEE Conference on
Decision and Control and European Control Conference.
Henningsson, T., Johannesson, E., and Cervin, A.
(2008). Sporadic event-based control of first-order linear
stochastic systems. Automatica, 44(11), 2890–2895.
Holmes, D. and Lipo, T. (2003). Pulse Width Modulation
for Power Converters: Principles and Practice. Wiley-
IEEE Press.
Imer, O. and Basar, T. (2005). Optimal estimation with
limited measurements. In Proceedings of the 44th IEEE
Conference on Decision and Control and European Con-
Keener, J. and Sneyd, J. (2008). Mathematical Physiology:
I: Cellular Physiology. Springer.
Lemmon, M. (2010). Event-triggered feedback in control,
Systems, 293–358. Springer.
Li, L., Lemmon, M., and Wang, X. (2010). Event-
10 triggered state estimation in vector linear processes.
In Proceedings of American Control Conference, 2138–
2143.
Meng, X. and Chen, T. (2012). Optimal sampling and
performance comparison of periodic and event based
impulse control. IEEE Transactions on Automatic
Control, 57(12), 3252 – 3259.
Meng, X., Wang, B., Chen, T., Darouach, M., et al. (2013).
Sensing and actuation strategies for event triggered
stochastic optimal control. In Proceedings of the 52nd
IEEE Conference on Decision and Control.
Miskowicz, M. (2006). Send-on-delta concept: an event-
based data reporting strategy. Sensors, 6(1), 49–63.
Oksendal, B. (2003). Stochastic Differential Equations: An
Introduction with Applications. Springer Verlag.
Rabi, M. (2006). Packet Based Inference and Control.
Ph.D. thesis, University of Maryland.
Rabi, M., Johansson, K., and Johansson, M. (2008). Opti-
mal stopping for event-triggered sensing and actuation.
In Proceedings of the 47th IEEE Conference onDecision
and Control, 3607–3612.
Rabi, M., Moustakides, G., and Baras, J. (2012). Adaptive
sampling for linear state estimation. SIAM Journal on
Control and Optimization, 50(2), 672–702.
Shi, D., Chen, T., and Shi, L. (2013). Event-triggered max-
imum likelihood state estimation. Automatica, 50(1),
247-254.
Sijs, J. and Lazar, M. (2011). Event based state estimation
with time synchronous updates. IEEE Transactions on
Automatic Control, 57(10), 2650–2655.
Suh, Y., Nguyen, V., and Ro, Y. (2007). Modified Kalman
filter for networked monitoring systems employing a
send-on-delta method. Automatica, 43(2), 332–338.
Tabuada, P. (2007). Event-triggered real-time scheduling
of stabilizing control tasks. IEEE Transactions on
Automatic Control, 52(9), 1680–1685.
Wang, B., Meng, X., and Chen, T. (2014). Event based
pulse-modulated control of linear stochastic systems.
IEEE Transactions on Automatic Control. In press.
Wang, X. and Lemmon, M. (2011). Event-triggering in dis-
tributed networked control systems. IEEE Transactions
on Automatic Control, 56(3), 586–601.
Wu, J., Jia, Q., Johansson, K., and Shi, L. (2013). Event-
based sensor data scheduling: Trade-off between commu-
nication rate and estimation quality. IEEE Transactions
on Automatic Control, 58(4), 1041–1046.
Xu, Y. and Hespanha, J. (2004). Optimal communication
logics in networked control systems. In Proceedings of
the 43rd IEEE Conference on Decision and Control, vol
4, 3527–3532.
You, K. and Xie, L. (2013). Kalman filtering with sched-
uled measurements. IEEE Transactions on Signal Pro-
cessing, 61, 1520–1530.
5513