Pilot_ Passive Device-free Indoor Localization
Pilot:Passive Device-free Indoor Localization Using Channel State Information
Jiang Xiao ,Kaishun Wu ??,Youwen Yi ,Lu Wang and Lionel M.Ni
Department of Computer Science and Engineering
Guangzhou HKUST Fok Ying Tung Research Institute,HKUST
?National Engineering Research Center of Digital Life
State-Province Joint Laboratory of Digital Home Interactive Applications,Sun Yat-sen University
?Corresponding Author
Abstract—Many emerging applications such as intruder detec-tion and border protection drive the fast increasing development of device-free passive(DfP)localization techniques.In this paper, we present Pilot,a Channel State Information(CSI)-based DfP indoor localization system in WLAN.Pilot design is motivated by the observations that PHY layer CSI is capable of capturing the environment variance due to frequency diversity of wideband channel,such that the position where the entity located can be uniquely identi?ed by monitoring the CSI feature pattern shift. Therefore,a“passive”radio map is constructed as prerequisite which include?ngerprints for entity located in some crucial reference positions,as well as clear environment.Unlike device-based approaches that directly percepts the current state of entities,the?rst challenge for DfP localization is to detect their appearance in the area of interest.To this end,we design an essential anomaly detection block as the localization trigger relying on the CSI feature shift when entity emerges.Afterwards, a probabilistic algorithm is proposed to match the abnormal CSI to the?ngerprint database to estimate the positions of potential existing entities.Finally,a data fusion block is developed to address the multiple entities localization challenge.We have implemented Pilot system with commercial IEEE802.11n NICs and evaluated the performance in two typical indoor scenarios. It is shown that our Pilot system can greatly outperform the corresponding best RSS-based scheme in terms of anomaly detection and localization accuracy.
Index Terms—Device-free Indoor localization,Channel State Information,RSS,Physical Layer.
I.I NTRODUCTION
Indoor location based services(LBSs)are becoming ubiqui-tous popular for providing people location-aware information. Advances have been made to enable the indoor LBS using RF-based technologies such as WLAN,wireless sensors and Radio-frequency identi?cation(RFID),etc.Most of these technologies share a common requirement that special devices like WiFi-enabled smartphones or RFID tags must be car-ried.However,as LBSs are bringing forth new expectations, such device-based approaches become ineligible for satisfying some emerging application demands.For example,exhibition galleries and shopping centers are expecting to support the pilferage prevention and missing people tracking services in a way that the visitors and customers do not need to carry on speci?c hardware.Important applications also exist in other indoor settings like hospitals,residences and places of entertainment.In hospital,health care providers need to grasp the distribution of location of the wandered patients associated without a device and quickly expand the relief operations. Also,people can?gure out the position of the intrusive indi-viduals in a resident district for safety precaution.Therefore,a device-free passive technique capable of detecting,positioning, and tracking entities neither carry any devices,nor participate actively in the localization process will be greatly helpful.
In general,the underlying technical challenges for designing a passive device-free indoor localization system are in two folds.First noted that device-based class can inherently obtain the knowledge of current status with a device attaching to the target and directly do localization.In contrast,the na-ture of device-free scheme requires implementing a similar functionality by detecting the occurrence of anomalous entity in the area of interest.Therefore,the?rst challenge that has to be addressed in order to enable the novel location-aware applications is anomaly detection problem,also known as hu-man/motion detection.Second,how to accomplish localization when a motion event of an entity has been detected serves as the new knotty problem.State-of-the-art researches[3],[8],[9] adopt radio signal strength(RSS)as the base modality in an attempt to overcome these dif?culties.However,we argue that the performance of device-free positioning systems based on RSS is limited by the disadvantage of RSS itself.Speci?cally, indispensable anomaly detection can be suffered from the high variability of RSS,owing to its coarse measurement.More-over,the inherent?uctuation of RSS makes it less sensitive to entity-caused environmental changes,not to precisely signify a location?ngerprint.Consequently,there is a pressing need to prompt a new modality superior to RSS for device-free indoor localization.
Fortunately,physical layer Channel State Information(CSI) from OFDM-based system promises new potential to over-come the above limitations of RSS.Previously,CSI has proved to be a reliable metric for locating the entity with WiFi-enabled device[24].Under this ground,we envision a future of leveraging CSI for passive device-free indoor localization. To this end,we start by investigating the feasibility of CSI-based device-free scheme.Based on preliminary experiments, we obtain two key observations.The primary one is that CSI
2013 IEEE 33rd International Conference on Distributed Computing Systems
Fig.1:Delay Pro?le in Different Environments.
is capable of detecting anomaly that affected by changes in the environment.This relies on the temporal stable feature of CSI that ensures the sensitivity of capturing the environmental variance owing to abnormal entities’(i.e.,human)occurrence and movement.The second insight stems from adequacy of CSI to differ a?xed location where the entity is present from all the other locations.Frequency diversity[11]of CSI allows it to re?ect the varying multipath re?ections due to entities’existence.In Figure1,the time domain delay pro?le obtained by inverse fast Fourier Transform(IFFT)of frequency domain CSI shows that an entity in different positions will change the multipath re?ections differently and result in different delay pro?les.Thus,CSI offers two major bene?ts when detecting the abnormal entities and serving as location?ngerprint.
In this paper,we design Pilot,a CSI-based P assive device-free i ndoor lo caliza t ion system.Our main idea is to leverage the bene?cial characteristics of CSI to monitor the abnormal appearance(anomaly or motion detection)and then to identify the location of entity.In particular,we design three blocks to enable the passive device-free localization functionality. First,we explore the frequency diversity of CSI in passive radio map construction block to generate normal and abnormal ?ngerprints.Second,anomaly detection block utilizes the correlation of CSI over time to monitor the abnormal variance. This block is the prerequisite of?nalizing locating the position of the anomaly entities,which is more challenge.Third,we tackle this knotty localization problem with position estimation block.Nevertheless,we develop a data fusion algorithm to determine the positions of multiple entities.
The main contributions of Pilot system are summarized as follows:
1)We exploit the feasibility of using?ne grained channel
state information for passive indoor localization.To the best of our knowledge,this is the?rst work to leverage PHY layer information CSI for DfP indoor localization in WLAN.
2)We take the advantages(temporal stability and frequency
diversity)of CSI to design Pilot,a passive indoor local-ization system,to realize passive radio map construction,
anomaly detection,and position estimation,respectively.
3)Extensive evaluations of Pilot with commercial802.11
NICs are conducted in two typical indoor scenarios.
These measurements show that the Pilot provides higher anomaly detection ratio than RSS-based RASID.Pilot, Pilot greatly outperforms RSS-based Nuzzer system with respective to localization accuracy.
The remainder of this paper is structured as follows.Sec-tion III discusses the central two observations that motivate our approach.Then we summarize the state-of-the-art researches on indoor localization in Section II.Section IV presents the overall architecture design of Pilot along with detailed methodology.In Section V,we describe the implementation of Pilot,and evaluate the performance in two typical indoor environments.Finally,we render our conclusions and present avenues for further research based on this work in Section VI.
II.R ELATED W ORK
Indoor localization has gained worldwide attention for its advantages of providing location awareness for various kinds of LBSs.There are primarily two categories of techniques related to this and become an increasing popular research?eld, namely device-based and device-free techniques.We expand upon representative prior studies in each of these two-fold techniques below.
Device-based techniques:Existing and emerging indoor lo-calization systems mainly depends on device-based techniques that targeted entities can only be localized with attaching a device.To name a few,LANDMARC[19]employ densely deployed RFID tags as receiver in the positioning region of interest;Cricket[27]and Active Badge[16]separately handle the localization problem by leveraging ultrasonic and infrared sensors;wireless sensors such as MicaZ[7]and TelosB[10]employed in various scale testbed enable an alter-native approach for location estimation;FM radio[17]is also proposed for positioning purpose.However,they all require a speci?c hardware to facilitate measurements for localization. In addition,some of them constrained to particular conditions such as infrared can function with the necessity of light of sight(LOS)existence.Alternatively,the pioneer RADAR[28] system investigates radio signal strength(RSS)to measure the distance between the APs and WiFi-enable receivers. Horus[20]improves the accuracy by applying a probabilistic model of RSS distribution.To avoid time-consuming site survey,WILL[23]augments user motions with the RF signal characterisers to construct a logical radio map for localization. In[4],[5],[6],the authors propose to use temporal channel response as link signature for differentiating locations.Even though these systems employ the already installed WLAN infrastructure without additional cost,they still require efforts from carrying on device at the transmitter that inappropriate for ubiquitous scale setting.On the contrary,our Pilot system is purely passive and device-free.
Device-free techniques:Driven by the necessity of satis-fying expectations of new kinds of location-aware services, device-free techniques[1]-do not require the entities to carry
any device-have gained widespread concern by research community.
Computer vision[18],[22]and RFID tags[9],[31]have been deployed in an indoor environment for device-free lo-calization functionalities.In a similar fashion,the authors propose a similar concept of“transceiver-free”[8],and use wireless sensor networks for building a RF-based object tracking system[21],[9].In[29],[22],Radio Tomographic Imaging(RTI)technique is presented for imaging the passive moving objects by applying a linear model.In[26],the authors makes improvement by leveraging motion-induced variance of RSS measurements.Yang et.al.[12]develop a joint learning GREEK algorithm to effectively diagnose the presence of intrusions in Zigbee network.However,the above approaches loss attraction in terms of scalability due to either high speci?c hardware cost like video camera and RFID tags and maintainable cost(i.e.,sensors).WLAN-based approaches[1], on the other hand,use the available infrastructure for indoor localization.Recently,the authors have developed a RSS-based Nuzzer system in a large scale indoor setting with pre-installed WLAN infrastructure[3].In this paper,we introduce the use of a new metric CSI from PHY layer for device-free indoor localization to replace the coarse RSS value,which can be resist to temporal variance and sensitive to environment changes by exploiting the frequency diversity.To the best of our knowledge,this is the?rst work to apply?ne-grained CSI to improve performance of device-free indoor localization in WLAN.
III.B ACKGROUND AND H YPOTHESES
A.PHY Layer Channel State Information
Our system leverages CSI value for device-free indoor localization.We therefore review such CSI value in this section.
In wireless communications,Channel State Information (CSI)is a?ne-grained PHY layer information that describes the channel property of a radio frequency(RF)link at the subcarrier level.To be more speci?cally,CSI describes how a RF signal propagates from the transmitter(s)to the receiver(s) and reveals the combined effect of,for instance,scattering, fading,and power decay with distance.Generally,CSI is a collection of M×1matrices H that speci?es channel gain over a pairs of transmitter and receiver with multiple antennas over M subcarriers.Mathematically,CSI on a single subcarrier can be represented by amplitude(|h|)and phase(∠h) as h=|h|e j sin{∠h}.Based on IEEE802.11n standard,the commercial wireless network interface card(NIC)allows us to obtain CSIs conveniently.
CSI based on OFDM system has gained popularity in a couple of applications.To name a few,authours in[30] propose to utilize CSI for rate adaptation instead of widely used RSS.That is,wireless packet delivery can be accurately predicted by using CSI.In[24],CSI is shown to be appropriate for device-based indoor localization.CSI offers the capability of estimating the distances between transmitters and receivers. In[25],we introduce the use of CSI for indoor motion detection,which can be sensitive to environment changes and
resist to temporal variance.In this paper,we further explore
the favorable features of CSI for realizing device-free indoor
positioning.
B.Hypotheses and Measurements
In this section,we start our work by testing two hypotheses
of utilizing CSI for device-free localization.We demonstrate
that,such hypotheses provide insight for our eventual system
design.Afterwards,based on preliminary measurements,we
validate these hypotheses and shed some light on the design
of a new CSI-based device-free localization system.
We present two integrant hypotheses of designing a CSI-
based device-free localization system as follows: Hypothesis1:CSI over multiple subcarriers can reveal
the abnormal status caused by appearance of human.More
speci?cally,a motion behavior of an entity will cause CSIs
variance that exhibits some kind of feature pattern shift.
To support typical device-free location-aware applications
such as intruder localization,the primitive step we need to
conduct is detecting the presence of a suspicious device-
free entity,i.e.,the motion.This motion behavior indicates
an abnormal event happened during the whole localization
process,termed as a“localization trigger”.Once a“trigger”
is detected,the location estimation block is followed and
?nalized.However,how to perform passive anomaly detection
to serve as“localization trigger”is challenging as the intrusive
entities usually do not carry any radios or may not even
cooperative.Recently,RSS has been used for device-free
motion detection in RASID system[2].However,the fatal?aw
of RSS lies in its susceptibility to measurement itself due to
severe multipath effect in indoor environment.For this reason,
we consider to study the feasibility of leveraging a more
reliable CSI for detecting the mobility of entity.The intuition
is to investigate whether the CSIs over multiple links can
appropriately infer a moving entity by extracting the feature
patterns.In our design,we?rst attempt to apply single pair
of AP and DP to monitor this“trigger”occurrence. Hypothesis2:CSI over multiple subcarriers can be lever-
aged to distinguish entity,i.e.,human,in different locations.
That is,CSIs show some kind of differential feature patterns
when an entity appears at different locations.
After a motion behavior is detected,next comes to identify
the location of the entity in the region of interest.In fact,the
presence of an entity will in?uence the RF links between APs
and DPs in a typical indoor environments.On the basis of the
location of entity,the two-way impacts can be classi?ed as:?Direct light-of-sight(LOS)blocking:the entity is located exactly between the AP and DP such that blocks the direct
LOS transmitting link;
?Indirect non-light-of-sight(NLOS)re?ection:the position of entity lies beside the LOS link and in?uences the multipath propagation of RF signals.
CSI is capable of revealing the change of channel status due
to the blockage of LOS path.In addition,it can present
multiple NLOS re?ection due to frequency diversity as shown
Fig.2:Anomaly Detection by CSI Feature Shift.
Fig.3:Location Distinction by Variant CSI Features.
in Figure1.Therefore,we hypothesize that the CSI will exhibit unique feature pattern at a given location that different from the others due to the impact of an entity’s motion behavior.In this way,a“passive”radio map can be constructed by storing the CSI over each RF link as a?ngerprint for each location. We verify these two hypotheses using the following prelim-inary experiments.
Experiment1:Anomaly Detection by CSI Feature Shift Our?rst experiment examines the effects on CSI when an anomalous entity is appeared in the monitoring region. We expect that CSI will exhibit distinguishable characteristics between static status and dynamic status.Under static status, we collect CSIs of n packets and store them into a passive radio map,namely normal?ngerprints database(DB)H Nor as:
H Nor=(H1,H2,...,H n)(1) We calculate the self-correlation of the CSIs of each packet i with all the rest packets and afterwards average the aggregated correlation sum as follow,
C i Nor=1
n
n
j=1
corr(H i Nor,H j Nor)(2)
This C i Nor is set as a“normal”correlation value for detecting an abnormal behavior.We set up an abnormal environment where only one person is present in the area of interest. Similarly,we measure the CSIs and compare them to the constructed normal pro?les by applying correlation function
as:
C i Abn=1
n
n
j=1
corr(H j Nor,H i Abn)(3)
Figure2plots the empirical probability density function (PDF)of CSIs feature patterns between these two kinds of statuses.Clearly,there is an obvious feature shift of CSI cor-relation when encountering an anomalous event.It means that CSI is temporal stable in static environment while sensitive enough for an instantaneous motion response.
Experiment2:Location Distinction by Variant CSI Features
Our next experiment is to inspect if the variant feature patterns of CSIs show uniqueness for a given location with entity and can be applied as?ngerprints for localization.
In Figure3,we depict the cumulative distributive function (CDF)of CSI feature pattern across6locations.To be speci?c, the red curve presents the CDF of self-correlation for position 1,while other5curves show the CDF of cross-correlation between each with position1,respectively.It is observed that one location can be distinguished from the others by analyzing the statistical properties of cross correlation.More concretely, these CSI feature patterns provide appropriate information regarding differentiating locations as?ngerprints.
In summary,we have made two important observations in this section:
1)CSI can capture the environment variance due to its
temporal stability;
2)CSI can differ a given location where the entity appears
from all the other locations.
This motivates us to apply CSI in device-free technique to achieve high localization accuracy.In what follows,we detailed design a novel CSI-based passive indoor localization system.
IV.T HE P ILOT D ESIGN
In this section,we present the design of Pilot system.We begin with an overview of Pilot architecture with three key constituent blocks.Then,we lay out detailed description of each block in subsequent subsections.
A.Overview
Pilot is built on the WLAN infrastructure without additional deployment and management cost.In our design,Pilot consists of three hardware elements:access points(APs),detecting points(DPs),and Pilot server as depicted in Figure4.A radio frequency(RF)link will be established between a pair of AP and DP kept stationary during the whole localization period. The AP broadcasts beacon message periodically.Besides involved in the activity of localization,those APs can also serve as hotspots simultaneously.The DP is a general wi?compatible device which is responsible for interacting with
Fig.4:Hardware elements of Pilot.
Fig.5:Pilot Architecture.
both AP and server.Upon receiving beacon messages from
the AP,the DP will record the according CSI.These raw CSIs
across multiple subcarriers at the PHY layer are then uploaded
to the Pilot server.Figure5shows the architecture of Pilot
system.Pilot server will perform localization by carrying out
three main procedures as below.
1)Passive Radio Map Construction Passive Radio Map
Construction block is primarily developed for two pur-
poses:1)to generate a normal CSI pro?le for anomaly
detection;2)to build up an abnormal database corre-
sponding to different positions where entity is located
for position estimation.It is worth mentioning that this
radio map is“passive”since its generation involves no
active participation with device-based entity.
2)Anomaly Detection We then design the Anomaly De-
tection block to capture environmental changes due to
entity’s appearance.Pilot continuously executes mon-
itoring of CSI feature variance that indicates whether
an entity emerges or not in the area of interest.Once
an abnormal event has been detected,it triggers the
immediately subsequent execution of localization.
3)Position Estimation The Position Estimation block is
designed to map the instant abnormal CSIs to passive
radio map,and fuse the data over multiple links.Such
that the exact location of an abnormal entity can come
to knowledge.
In the following subsections,we will describe each block
of Pilot in a divide-and-conquer manner.
B.Passive Radio Map Construction
First,we propose an of?ine block-Passive Radio Map
Construction-as a basis to facilitate subsequent operations in
?ngerprinting system Pilot.Map construction block consists of
two functions:processing the measurement data and generate
the?ngerprints database.In OFDM-based networks,a RF
signal is transmitted over multiple subcarriers simultaneously
Fig.6:CSI correlation Gaussian Hypothesis test.
occupied a wide band.In a typical indoor setting,the chan-
nel is affected by frequency-selective fading and frequency
independent attenuation.The coarse RSS fails to uncover the
channel state at subcarrier level.In comparison,signi?cant
diversity over heterogeneous subcarriers can be captured by
CSI.Therefore,the underlying idea is to exploit the frequency
diversity of CSI,which reveals diverse feature patterns of dif-
ferent locations due to appearance and movement of abnormal
entities.
To begin with,we collect CSIs samples over the RF links
between APs and DPs in a normal state,i.e.,with no entity
appearance.We then process these sample data received from
multiple DPs on Pilot server.A normal?ngerprints database
denoted as C Nor composes of a set of n processed CSI
samples will be constructed.As stated in previous section,the
process of the normal pro?le construction is to average the
sum of the self-correlation of CSI samples over each RF link.
We use QQ?plot to test the Gaussian distribution hypothesis
of C Nor,and Figure6shows that it doesn’t pass Gaussian
hypothesis test,nor tests for other well-known distributions.
Fig.7:Sensing Zone and Dead Spot.
Note that the calibrated normal pro?le is a premise for
anomaly detection.In what follows,we need to generate a passive radio map in an of?ine phase for indoor localization.Unlike map construction of device-based ?ngerprinting sys-tem,this passive radio map construction is conducted with entities located at different positions without any devices,and the CSIs are collected by the DPs over each RF link.Similarly,the self-correlation of the CSIs for each position is stored in the database.
With the objective of narrowing the ?ngerprint mapping range,and also improving the localization accuracy with single link,we introduce two novel concepts as following:De?nition 1.Sensing Zone is the zone where the present of entity can be detected by a speci?c AP-DP link.
De?nition 2.Dead Spot is the position in the area of interest that outside the sensing zone of any AP-DP link.
Intuitively,the further the entities away from the monitor link,the less in?uence its appearance on the CSI measure-ments.Hereby,“Sensing Zone ”is christened to capture the in?uence of abnormal entities on RF links as shown in Figure 7.If the entity is positioned in the “Dead Spot ”(i.e.,outside the sensing zone),the server will not be able to distinguish this situation between normal one,and results in false negative.
Note that,due to the multipath re?ection and LOS blocking of signal in indoor environment,the shape of the Sensing Zone of a transceiver link is not a regular one like circle or ellipse,but much more complicated.Therefore,the Sensing Zone can only be characterized with ?eld measurements.In this work,we determine the Sensing Zone with CSI correlation feature for each link as following.We calculate the cross correlation of CSIs on each position in the passive radio map and the normal CSIs for each link according to Eq.(3).Then we will check the density distributions of C Nor and C Abn are signi?cantly different.If the difference is big enough,the position is within the Sensing Zone of this RF link.Since the distribution is not
Gaussian or well-approximated by other known distributions,we can use nonparametric hypothesis test.Speci?cally,we perform Ansari-Bradley test [13]where the null hypothesis de?ned as:
H 0:C Nor and C Abn follow the same distribution.If H 0of identical distributions cannot be rejected at the α%signi?cance level,we can conclude that they follow the same distribution,and the corresponding position is outside the sensing zone of the speci?c link.With these two concepts,we can reduce the searching space when motion is detected by a speci?c RF link.In addition,we can also optimize the positions of transmitters and receivers,so that all the monitor spots in the area of interest can be covered with minimal deployment cost.C.Anomaly Detection
A process of determining abnormal events from CSI mea-surements is a prerequisite for device-free indoor localization.Anomaly Detection block is designated to continuously ex-ecute this process.As mentioned in Section III ,we observe that CSI stays relatively stable over time in a normal state (i.e.,static)without entities’appearance or movement.Whereas in an abnormal status (i.e.,mobile),CSI experiences a feature pattern shift.Therefore,the basic idea is to leverage the temporal stability characteristic of CSI consistent with normal status so as to distinguish from the feature patterns under abnormal environments.In Pilot,we utilize this different feature pattern shift as a “localization trigger”for deciding whether a localization process should be started.
The main idea is to check the probability of each CSI feature to be in normal pro?le.To decrease the effect of outliers,we use a sliding window to average those raw CSI measurements.Now we attempt to estimate the probability of the current CSI correlation sample C according to the statistics of C Nor .However,the distribution of normal pro?le cannot be assumed as Gaussian and an alternative method for distribution estimation is in need.Therefore,we alternatively adopt a kernel density-based function approach as explained below.
In statistics,this approach is known as kernel density estimation (KDE).The bene?t of KDE is that it can estimate the density directly from the data without assuming a particular form for the underlying distribution.For RF link l between a pair of AP and DP,we denote W as the sliding window length,and n as the total number of correlation samples in the normal pro?le.Now consider a sequence of independent and identically distributed (i.i.d.)random correlation samples (C 1,C 2,...,C M )of M +W ?1packets.In our method,we de?ne the kernel density estimator as ?f l ,
?f l (C )=
1nh l n j =1K(C ?C l,j Nor h l
)(4)
where K is the kernel function and h l is the bandwidth.In particular,Epanechnikov quadratic kernel [14]is chosen
owing to ensure the fairness of comparison to the counterpart RSS-based approach [2]and given by:
K (u )= 3
4(1?u 2),if |u |≤1
0,otherwise (5)
h l is a scaling factor that controls how wide the probability mass is spread around a point as well controls the smoothness or roughness of a density estimate.According to Scott’s rule [15],the optimal bandwidth is given by:
h ?l =2.345?
σl n ?0.2(6)where ?σl is an estimate for the standard deviation for
C l,j Nor .
For each RF link,we examine the cumulative distribution
function (CDF)of the sample CSI correlation as ?F
(C Nor ).It should be noted that C Abn is always smaller than C Nor .Thus,the anomaly detection problem is equivalent to determine
whether C is smaller than a lower bound ?F
?1(β)determined by a preset value β.The selection of βprovides a tradeoff between false alarm and miss detection.
D.Position Estimation
In the previous two blocks,Pilot of?inely maintains a set of ?ngerprints into an abnormal radio map and continuously monitors the anomaly event in an online phase.As the target of locating the entities in real time,we now introduce the Position Estimation block.
Our objective is to accurately map the current ?ngerprint of the abnormal entity to the passive radio map during the online phase.The overall idea is to compare the obtained CSI measurements against the abnormal passive ?ngerprints database and thus selects the best match.Pilot chooses the maximum a priori probability (MAP)algorithm,which is a well-known probabilistic algorithm for performing ?ngerprint-based position estimation.
During the online localization stage,for an unknown lo-cation L where an abnormal entity is presented,we col-lect the abnormal CSI measurement H l Abn from link l .Let L =L 1,L 2,···,L m be the set of m locations on the passive radio map.Then,our position estimation task equates to ?nd a location L ∈L that maximizes a priori probability P (H l Abn |L ).On the basis of Bayes’law,we formulate this optimization object function by:
L ?
=arg max L
P (L |H l Abn )=arg max L [
P (H l Abn |L )P (L )P (H l Abn
)](7)Assume that all locations are equally probable,and P (H l
Abn )is independent of location L ,we have
L ?=arg max L
P (H l Abn |L )
(8)
Since H l Abn is high-dimensional (52subcarriers out of total
64in 802.11n standard)and partially correlated,the statistical analysis of H l Abn is very complicated.Therefore,similar to anomaly detection,we consider the cross correlation of H l Abn over the ?ngerprints at each position L ,and denote it as C l Abn,L .We use kernel density to represent the probability in Eq.(9)whose calculation is similar to that in Eq.(4).
E.Data Fusion
If the detection points are deployed with high density,it is possible that a single motion can be detected by multiple links as shown in Figure 7.Therefore,we can extend our basic scheme to such scenarios in order to further improve the localization accuracy with data fusion method.
Given the measurements of multiple RF links,the ?nal es-timation is the position that the joint possibility is maximized,i.e.,L ?=arg max L
l
P (H l Abn |L )
(9)To reduce the computational complexity,only links with
motion being detected will be included in the above calcula-tion.Moreover,only the positions in the common area of the sensing zone of these links will be selected as the candidate for ?ngerprint mapping.
In summary,given the abnormal CSI measurement at each RF link,the kernel density-based MAP algorithm outputs the location L with maximal kernel density.Note that,we only consider the scenario when a single intruder exists,the local-ization of multiple intruders will be much more complicated and thus beyond the scope of this paper.
V.P ERFORMANCE E VALUATION
In this section,we implement and evaluate Pilot in two typical indoor scenarios.We ?rst describe our testbeds and data collection methodology in Section V-A .Afterwards,we validate the performance of anomaly detection and localization in Pilot,along with the comparison against state-of-art RSS-based approaches.
A.Implementation
I.Experimental Scenarios:We set up two typical indoor testbeds in Hong Kong University of Science and Technology as listed below:
1)Laboratory First,we performed experiments in a re-search laboratory covers an area of 7m ×11m .It is surrounded by various of?ce facilities such as shelf,desk and chair,and therefore is subject to multipath effects.A total of 2pairs of DPs and APs are placed according to the ?oor plan in Figure 8.
2)Lobby Second,we deployed Pilot a larger testbed in an “L”-shape lobby,which spreads over approximately 776m 2.In this experimentation site,we symmetrically place two pairs of APs and DPs in opposite sides of parallelogram.The area of lobby is separated into a couple of squares and we choose 30reference positions,each of them are 4m apart as shown in Figure 9.
II.Data Collection:In our experiments,we use TL-WR941ND router as wireless AP that transmit information over a RF link to DPs.Pilot DP is a standard HP laptop equipped with commercial 802.11n 5300NICs,and the op-erating system is the Linux kernel 2.6.34.The Pilot server is running on one of these laptops.In our current implementation,we only use the ?rst antenna and the enhancement with multiple antennas is left for our future work.
Fig.8:Fingerprint Layout in Laboratory.
Fig.9:Fingerprint Layout in Lobby.
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
D
e
t
e
c
t
i
o
n
R
a
t
e
Pilot
RASID?like
Fig.10:Accuracy of Anomaly Detection.
During the whole localization period,APs will continuously
send out beacon messages to DPs.DPs gather these messages
along with CSIs and upload them to detection server for
processing.In each scenario,we?rst collect measurement
of every link in static environment without any person in
the area,and construct normal pro?le.Then,we divide the
geographic area of interest into uniform square grids,and mark
some of them as crucial reference positions.Passive abnormal
?ngerprints are collected when one volunteer stand on each
reference position.In this way,we construct passive radio
maps in both lab and lobby.
B.Accuracy of Anomaly Detection
High accuracy of anomaly detection is necessary to guar-
antee the ef?ciency of device-free localization.In this section
we evaluate whether Pilot can achieve this goal and conduct
a comparison with the best RSS-based device-free motion
detection system RASID[2].
Figure10plots the anomaly detection results in lab and lob-
by using an Receiver Operating Characteristic(ROC)curve.
This ROC curve graphically reveals the inherent tradeoff
between the false positive(FP)rate and detection rate.FP
rate(X-axis)also known as false alarm,represents the pro-
portion that normal state is falsely detected as an anomaly.As
mentioned in Section IV-C,the parameterβplays an important
role in striking a balance of high detection rate with respect to
low FP rate.In our experiment,we adaptβand?x the sliding
window length to be10.From the Lab scenario Figure10(a),
we have two obvious observations:1)for a FP rate less than
or equal to10%,the detection rate of both approaches is alike
to be around40%;2)for a FP rate greater than or equal to
30%,the detection rate of Pilot is about90%that far exceeds
RSS-based RASID.We have similar observations in the lobby
scenario as shown in Figure10(b).These results con?rm that
the CSI-based Pilot is superior to RSS-based approach in terms
of anomaly detection.
C.Accuracy of Localization
So far,we have described the performance of anomaly
detection in a typical laboratory scenario.Clearly,a foremost
criteria–localization accuracy is left for discussion in the
following sections.
We analyze the location distinction accuracy of Pilot when
there is only one single abnormal entity in Lab.In this scenari-
Fig.11:Single Entity Localization Accuracy with Different Link Numbers
Fig.12:Single Entity Localization Accuracy with Different Sample Numbers
o,Pilot deployment composes of2pairs of APs and DPs and4 RF links in total.We evaluate the effect of increasing numbers of links(from1to4)on the performance of the approach proposed in Pilot and the best RSS-based Nuzzer[3].Instead of Gaussian distribution assumption,we use kernel density function to depict the distribution of RSS.Since kernel density is a better distribution approximation,the RSS-based system is Nuzzer-like but improved one.These results are presented in Figure11(a),such that if only1link is available in the area,Pilot achieves higher accuracy(6%)than Nuzzer-like approach.As changing the links from2to4,the accuracy of Pilot can extend to over90%even to98%while Nuzzer-like approach only has a slight improvement.In other words,the use of more RF links would lead to an obvious increase in location distinction accuracy of CSI-based Pilot rather than RSS-based Nuzzer.This indicates the bene?ts of Pilot over RSS-based method in gaining distinction capability.We also study the location distinction performance of a single abnormal entity in Lobby.Similar performance is achieved with similar deployment as shown in Figure11(b).Comparing Figure11(a) and Figure11(b),we can observe that our system performs better in Lab scenario due to the more abundant multipath re?ections.
Figure12(a)is a result in Lab that compares the preci-sion of Pilot and Nuzzer with respect to different numbers of samples collected as?ngerprints.This?gure shows that increasing sample number will bring in higher accuracy in some extent,because with more?ngerprints,we can better approximate the distribution of CSI correlation,and results in lower?ngerprint mapping error according to(8).More than10%percents accuracy improvement can be obtained with more samples in our experiment.However,more samples means longer of?ine phase is required for map construction. Therefore,there is a tradeoff between the data collection time and localization accuracy.It is also shown that the proposed CSI-based Pilot always outperforms the corresponding RSS-based Nuzzer with respect to localization accuracy.Similar performance is achieved in the Lobby scenario as shown in Figure12(b).
From empirical experiments in these two scenarios,we can conclude that frequency diversity of CSI helps Pilot outperform the RSS-based scheme and such advantage is obvious when more RF links are available.
VI.C ONCLUSIONS AND F UTURE W ORK
In this paper,we present Pilot,a CSI-based passive device-free indoor?ngerprinting system in WLAN.It is the?rst proposal to leverage temporal stability and frequency diversity characteristics of CSI for developing a“passive”?ngerprint for device-free localization.In contrast to the traditional?n-gerprint approach,we integrate an Anomaly Detection block to facilitate the device-free nature.We apply the kernel density-based approach to calculate the CSI correlation and thus detect abnormal entities.Moreover,we develop another block for estimating the location of target,and analysis the feasibility of distinguish multiple entities simultaneously.We implement Pilot with commercial IEEE802.11n NICs and evaluated its performance in two different indoor scenarios.Pilot outper-forms the RSS-based RASID system on anomaly detection and Nuzzer system on localization with the same infrastructure de-ployment.The high accuracy of Pilot system demonstrates its potential to dramatically improve the performance of existing location-dependent applications.
In the perspective of future research,we intend to ex-ploit other approaches other than?ngerprinting for CSI-based device-free indoor localization to of?oad the calibration effort-s,and improve accuracy as well.Moreover,multiple antennas can also be leveraged to achieve better system performance.
A CKNOWLEDGMENT
This research is supported in part by Guangdong Nat-ural Science Funds for Distinguished Young Scholar(No. S20120011468),Hong Kong RGC Grant HKUST617212, Guangzhou Pearl River New Star Technology Train-ing Project(No.2012J2200081),Guangdong NSF Grant (No.S2012010010427),China NSFC Grant61202454.
R EFERENCES
[1]M.Youssef,M.Mah, A.Agrawala,“Challenges:Device-free Passive
Localization for Wireless Environments,”in Proc.of ACM MobiCom, 2007.
[2] A.E.Kosba,A.Saeed,M.Youssef,“RASID:A Robust WLAN Device-
free Passive Motion Detection System,”in Proc.of IEEE PerCom,2012..
[3]M.Seifeldin,A.Saeed,A.E.Kosha,A.EI-Keyi and M.Youssef,“Nuzzer
A Large-Scale Device-Free Passive Localization System for Wireless
Environments,”in Proc.of IEEE Transactions on Mobile Computing, 2011.
[4]N.Patwari,“Robust Location Distinction using Temporal Link Signa-
tures,”in Proc.of ACM MobiCom,2007.
[5]J.Zhang,M.H.Firooz,N.Patwari,S.K.Kasera,“Advancing Wireless Link
Signatures for Location Distinction,”in Proc.of ACM MobiCom,2008.
[6]J.Zhang,S.K.Kasera,N.Patwari,P.Rai,“Distinguishing Locations
Across Perimeters Using Wireless Link Measurements,”in Proc.of IEEE INFOCOM,2011.
[7]Z.G.Zhong,D.Wang,and T.He,“Sensor Node Localization Using
Uncontrolled Events,”in Proc.of IEEE ICDCS,2008.
[8] D.Zhang,J.Ma,Q.Chen,and L.M.Ni,“An RF-Based System for
Tracking Transceiver-Free Objects,”in Proc.of IEEE PerCom,2007.[9] D.Zhang,Y.H.Liu,L.M.Ni,“RASS:A real-time,accurate and scalable
system for tracking transceiver-free objects,”in Proc.of IEEE PerCom, 2011.
[10]X.N.Guo,D.Zhang,Lionel M.Ni,“Localizing Multiple Objects in an
RF-based Dynamic Environment,”in Proc.of IEEE ICDCS,2012. [11] A.Bhartia,Y.Chen,S.Rallapalli,and L.Qiu,“Harnessing Frequency
Diversity in Wi-Fi Networks,”in Proc.of ACM MobiCom,2011. [12]J.Yang,Y.Ge,H.Xiong,Y.Y.Chen,and H.B.Liu,“Performing joint
learning for passive intrusion detection in pervasive wireless environ-ments,”in Proc.of IEEE INFOCOM,2010.
[13] A.R.Ansari and R.A.Bradley,“Rank-sum test for dispersions,”in
Proc.of Ann.Math.Statist,1174-1189,1960.
[14]Epanechnikov,V.A.,“Non-parametric estimation of a multivariate prob-
ability density,”in Proc.of Theory of Probability and its Applications, 1969.
[15] D.W.Scott,“Multivariate Density Estimation:Theory,Practice,and
Visualization,”,in Proc.of John Wiley&Sons,Inc.,1992.
[16]Want,R.and Hopper,A.and Falc?a o,V.and Gibbons,J.,“The active
badge location system,”in Proc.of ACM Transactions on Informaton Systems,1992.
[17]Y.Chen,D.Lymberopoulos,J.Liu,B.Priyantha,“FM-based Indoor
Localization,”in Proc.of ACM MobiSys,2012.
[18]J.Krumm,L.Williams,and G.Smith,“SmartMoveX on a Graph-An
Inexpensive Active Badge Tracker,”in Proc.of UbiComp,2002. [19]Ni,L.M.and Liu,Y.and Lau,Y.C.and Patil,A.,“LANDMARC:indoor
location sensing using active RFID,”in Proc.of IEEE PerCom,2003.
[20]Youssef,M.and Agrawala,A.,“The Horus WLAN location determina-
tion system,”in Proc.of ACM MobiSys,2005.
[21] D.Zhang,L.M.Ni,“Dynamic Clustering for Tracking Multiple
Transceiver-free Objects,”in Proc.of ACM SenSys,2009.
[22]Y.Zhao and N.Patwari,“Noise Reduction for Variance-Based Device-
Free Localization and Tracking,”in Proc.of IEEE SECON,2007. [23] C.Wu,Z.Yang,Y.H.Liu,and W.Xi,“WILL:Wireless Indoor
Localization without Site Survey,”in Proc.of IEEE INFOCOM,2012.
[24]K.Wu,J.Xiao,Y.Yi,and Lionel M.Ni,“FILA:Fine-grained Indoor
Localization,”in Proc.of IEEE INFOCOM,2012.
[25]J.Xiao,K.Wu,Y.Yi,and Lionel M.Ni,“FIMD:Fine-grained Device-
free Motion Detection,”in Proc.of IEEE ICPADS,2012.
[26]J.Wilson and N.Patwari,“See-Through Walls:Motion Tracking Using
Variance-Based Radio Tomography Networks,”in Proc.of IEEE Trans-actions on Mobile Computing,2011.
[27]Priyantha,N.B.and Chakraborty,A.and Balakrishnan,H.,“The Cricket
location-support system,”in Proc.of MobiCom,2000.
[28]P.Bahl and V.N.Padmanabhan,“RADAR:An inbuilding RF-based
user location and tracking system,”in Proc.of IEEE INFOCOM,2000.
[29]J.Wilson and N.Patwari,“Radio Tomographic Imaging with Wireless
Networks,”in Proc.of IEEE Transactions on Mobile Computing,2010.
[30]Halperin,D.and Hu,W.J.and Sheth,A.and Wetherall,D.,“Predictable
802.11packet delivery from wireless channel measurements,”in Proc.of ACM SIGCOMM,2010.
[31]R.S.Moore,R.Howard,P.Kuksa,and R.P.Martin,“A Geometric
Approach to Device-Free Motion Localization Using Signal Strength,”in Technical Report,Rutgers University,2010.
用跟造句
你可以如同以海为家的水手们那样，视它们为你的指引，跟随着它们，最后必能达到你的目的地。 11、行动好像是跟随感觉走的，其实不是如此，行动是与感觉 平行的，我们能使直接受意志支配的行动有规律，也能间接的使不 直接受意志制约的支配有规律。 12、幸福的理解有千万种，每人的诠释都不同，但是人生最大 的幸福就是可以做自己。相信自己，跟随自己的心灵和直觉，不要 盲从信条，不要盲目攀比，你就会是最幸福的。 13、声誉有聪明的，也有愚蠢的；有公正的，也有不公正的； 声誉有短促的、轻率的、昙花一现的；也有缓慢的、艰难的、紧跟 着创造后面羞怯而来的。有的声誉凶多吉少，总是姗姗来迟，并耗 尽了人的心血。 14、跟随而来的是欢迎新人的夹夹乐，合作不错啊！最后是抱 抱团，大家欢声笑语一片！ 15、幸福？人们跟随它，追寻它，人们为此累得要死，而它并 不在那里，它只产生于人们不再有任何期待、不再对任何事物有所 怀疑的时候。 16、如果你要成功，你应该朝新的道路前进，不要跟随被踩烂 了的成功之路。 17、每个人的生命都是由两种不同方向的矛盾重重的力量支配。挣扎是来自于你跟随向上的力量还是向下的力量。但你必须要保持 自己内心的纯洁。愉悦与坚定，而不管外界环境如何。 18、“小暴”是我们的班主任，她的名字叫暴丽丽，老师都叫 她“小暴”，所以我们也跟着叫开了“小暴”。你们别认为暴老师 姓“暴”，就认为她脾气暴躁，很容易发火。其实“小暴”是很热
爱同学们的，她一般不发火。 19、如果人生是一段旅途，快乐与悲伤就是那两条长长的铁轨，在我身后紧紧跟随。 20、恋爱不过是一场高烧、思念是紧跟着的好不了的咳嗽。 21、远处是重重叠叠、连绵不断的山峰，山峰青得象透明的水晶，可又不那么沉静。我们的车子奔跑着，远山也象一起一伏的跟 着赛跑；有时在群峰之上，又露出一座更秀隽的山峰，象忽地昂起 头来，窥探一下，看谁跑得快。 22、你看它虽然迷了路，仍傲然地前进着。它不断地左冲右撞，终于走出一条路。我的目光跟着它的脚步，它走着，走着，一路上 遇到不少同伴，它们互相打着招呼。 23、学问是我们随身的财产，我们自己在什么地方，我们的学 问也跟着我们在一起。 24、毽子忽左忽右忽上忽下，在姗姗的身前身后蹦来蹦去，姗 姗时而转向这边，时而转向那边，紧紧地跟着毽子，眼睛始终随着 毽子灵活地转动着。 25、朋友是一直默默跟随身后的影子，当整个世界离你而去时，才发现只有他坚定不移。 26、一会儿，乌云弟弟不知从哪里钻了出来，使天空变得阴沉 沉的，好像快要塌下来。电婆婆也到了，挥起利剑在天空划出无数 的光芒。雷公公也紧跟着凑热闹，不客气地敲起大鼓，轰隆隆！轰 隆隆！让人不寒而栗。 27、春天农民辛辛苦苦的插种，播种等的工作，夏天又要杀虫，秋天又要忙着收获，跟着再种别的农作物。一年四季，这样的顺序 重重复复，为的就是种出蔬菜，然后卖到批发市，再到市场，最后
用尽管还是造句
用尽管还是造句 1. 永远不要无休止的围着你爱好的那个男人转，尽管你喜欢得他快要掏心掏肺的逝世掉了，也还是要学着给他空间，否则，你要警惕缠得太紧会勒逝世他的。 2. 永远不要无休止的围着你喜欢的那个男人转，尽管你喜欢得他快要掏心掏肺的死掉了，也还是要学着给他空间，否则，你要小心缠得太紧勒死了他。 3. 女孩那根直撅撅的像把钢刷子似的小辫，尽管用皮筋一道又一道紧紧地箍住，可还是挑战似的翘在后脑勺上。 4. 永远不要无休止的围着你喜欢的那个男人转，尽管你喜欢得他快要掏心掏肺的死掉了，也还是要学着给他空间，否则，你要小心缠得太紧会勒死他
的。 5. 闷热的下午，一丝风也没有，尽管我坐在树荫下，但还是热：脸是热热的，胸膛是热热的，连吸进的空气都是热热的，热得透不过气来。 6. 永远不要无休止的围着你喜欢的那个女人转，尽管你喜欢得她快要掏心掏肺的死掉了，也还是要学着给她空间，否则，你要小心缠得太紧她会反感。 7. 尽管我们用判断力思考问题，但最终解决问题的还是意志，而不才智。 8. 尽管大家都这么说,他还是半信半疑。 9. 生命是短促的，然而尽管如此，人们还是有时间讲究礼仪。爱默生 10. 我想说，实际我不像你们想象的那样很成功，尽管你发现我很自信，但实际上我还是一个普通人。王石 11. 人往往被爱所欺，为爱所伤，甚至陷入不幸。尽管如此，还是对人充满爱恋。 12. 尽管蚊子吵得很厉害，人们还
是鼾声如雷。 13. 我就觉得自己说了老实话反而毁了我自己；不过，尽管那样，您还是会看得起我的，这是我在世上最值得重视的事。大仲马 14. 尽管荆棘塞途,他们还是很顺利地完成了任务。 15. 尽管人生那么无情，我们本人还是应当把自己尽量改好，少给别人一些痛苦，多给人一些快乐。 16. 尽管我们用判断力思考问题，但最终解决问题的还是意志，而不是才智。 17. 尽管工作任务很忙，他还是见缝插针，坚持自学了大学英语。 18. 真正的信仰就是你在此刻正坚信的那一个，尽管它可能还是如此的错误。唯一重要的是，这是你的自己的信仰。 19. 尽管敌机在上空盘旋,他还是不动声色地筹划着作战方案。 20. 尽管小红已经连续好几次被
只要怎么造句
只要怎么造句 本文是关于只要怎么造句，感谢您的阅读！ 造句先把要造句的词扩展成词组，然后再把句子补充完整。只要怎么造句？下面一起来看看。 只要怎么造句 1、只要全班同学齐心，没有克服不了的困难。 2、无论做什么工作，只要对社会有贡献，就有出息。 3、个人利益只要是合法的，就应该得到保护。 4、只要大家努力，提前完成任务是可能的。 5、青少年只要努力学习，都有美好的前程。 6、只要坚持练，日久天长，定能写出一手好字。 7、只要下功夫，外语是可以学好的。 8、只要坚持体育锻炼，身体就会逐渐强壮起来。 9、只要你能自圆其说就行，不必那么标准。 10、电子计算机并不神秘，只要认真学就能掌握它。 11、夜间走路，只要认出北极星，就能辨别出东西南北。 12、人犯错误不要紧，只要知错即改就是好同志。 13、只要是合理的要求，妈妈总会满足我的。 14、无论从事什么工作，只要努力，都有前途。 15、俗话说，只要工夫深，铁杵磨成针。 16、古人早就说过言者无罪，闻者足戒，大家有什么意见，只要提出来，就是对我的帮助。
17、只要他一抛头露面，我们便能发现。 18、我虽然力量有限，但只要肯接受我的帮助，也许不无小补。 19、只要他说出实话，我们还佛眼相看。 20、只要有一个较好的工作条件，即使恶衣恶食，又有什么关系。 21、赵家奶奶平常烧香拜佛，信奉好生之德，只要看见邻家杀鸡，也会双手合十，口称“阿弥陀佛”。 22、只要因风吹火，办这件事便花不了多大力气。 23、只要我们有实力，别人自然会另眼相看。 24、你只要知过必改，一定会成为一个好同志。 25、你不要灰心，只要改过迁善，前途仍然是光明的。 26、只要干部廉洁奉公，坏分子便无隙可乘。 27、要学好一门语言很不容易，但只要锲而不舍，总会有收获的。 28、社会要重视对青少年进行传统的道德教育，只要工作做得细，一定能使浪子回头。 29、干事业只要下定决心，任何艰难困苦都能克服。 30、对于当过几年差的旧军政人员，只要留下来的，我们便应该弃如弁髦，给他们一定的出路。 只要就造句 1、只要人人都献出一点爱，世界就会变得更美好。 2、只要努力学习就能获得知识同时也能考取好的大学。 3、只要努力学习定会取得优异成绩。 4、只要我多做些力所能及家务事妈妈会少我操心点。
三相多功能电力仪表
三相多功能电力仪表 （版本号：4.00） 使 用 说 明 书 西安天立电气科技有限公司（使用前请详细阅读此说明书）
目录 1、简述 (1) 1.1 的功能 (1) 1.2 功能描述 (1) 1.3 的特点 (1) 2、安装、接线与配置 (2) 2.1 尺寸与安装 (2) 2.2 接线与配置 (3) 3、界面显示说明 (5) 3.1 显示模式下按键功能示意图 (5) 3.2 编程模式下参数查询及修改按键功能示意图 (5) 3.3 显示模式下功能显示灯指示说明 (6) 4、操作说明 (6) 4.1 循环显示说明 (6) 4.2 键盘编程说明 (6) 4.3 功能设置 (6) 5、运输与贮藏 (7) 6、保修期限及订货说明 (7) 附表：常见故障排除 (8)
1、简述 1.1功能 三相多功能电力仪表是一款用于低压电力系统的智能化装置，它集数据采集和控制功能于一身，具有电力参数测量及电能计量为一体，其精度符合国家电能表标准，同时具有选配4路开关量输入、2路脉冲输出、RS485通讯接口，通讯协议为MODBUS 通讯协议或其它指定通讯协议。 1.2 功能描述 1.2.2 通讯功能（扩展功能） 自带RS485通讯接口，通讯协议为MODBUS通讯协议或其它指定通讯协议。可通过通讯口，可查询全部的测量监控信息。面板带有带有LED指示灯，用于通讯收/发指示。 1.2.3 1~4路开关量输入功能（扩展功能） 可选配1~4路开关量输入，为无源节点输入。可在测量显示区查看开关量的状态。 1.2.4 模拟量输出功能（扩展功能） 可选配1~2路DC4~20mA，通过编程设置可将模拟量（DC 4-20mA）输出设置为与某一被测参数(定货时需指定)成比例的输出。模拟量输出的最大负载为300Ω，可选择项为IA、IB、IC、UA、UB、UC。 1.2.5 脉冲输出功能（扩展功能） 可选配2路脉冲输出（有功脉冲输出/无功脉冲输出）。 1.3 的特点 -1- 1.3.1 安全性高，可靠性好 在设计过程中采用了多种抗干扰措施，能够在电力系统中稳定运行。静电放电抗
用像什么又像什么造句大全
用像什么又像什么造句大全 1、秋雨像圆圆的珍珠，又像滑润的碎玉儿，零零散散，断断续续。 2、往远处看，荷花像倾在湖里的胭脂，又像落在湖上的云霞，更是别具一格。吟诵着那“接天莲叶无穷碧，映日荷花别样红”的诗句，顿觉心旷神怡，浮想联翩。 3、在发凉的地板上静静坐着，听着掠过窗外的风，恍惚间像是小儿的梦呓，像是情人的低语，又像是临行前母亲的叮咛。 4、蓝天上挂着一轮火红的太阳，就像一只红彤彤的苹果，又像一只火红的灯笼。 5、她画的这幅画，好像画的是猫，又像画的是狗。 6、母爱的感觉就像河水那样温柔，又像高山那样严厉。 7、紫盈盈的葡萄像一颗颗珍珠，又像一颗颗玛瑙。 8、小青松的叶子四季青，像一枚枚绿色的小针，又像小刺猬身上坚硬的刺。别看它的体积小，它的用处可大了！这针状的叶子，细细长长的，不仅能储存水分，而且还能减少水份的蒸发，从而才能使松树四季常青。 9、渐渐的，渐渐的太阳从东方升起，晨雾一片一片的消失了，太阳公公露出欣喜的笑容，花朵也绽开了，那娇小的花瓣。小草叶子的露珠像一块块宝石，又像一颗颗星星，更像一粒粒金沙，这场景真是太美妙了！ 10、漫天飞舞的雪花像一只只美丽的白蝴蝶，纯洁无暇，晶莹剔
透，纷纷扬扬，飘飘洒洒地从一望无际的天空中飘落下来，它时而像一位隐士，时而像一只活泼的小精灵，时而又像一位优雅的公主，和我一起嬉戏玩耍。 11、繁茂的大树下，有一个清澈的小溪，挨挨挤挤的荷叶像一个绿色的盘子，又像一把碧绿的伞，荷叶给小溪给覆盖了，小溪像换了件翠绿的衣裳。还有几朵五颜六色的荷花笑开了脸。小溪美丽极了。 12、有的**铺在山坡上，有的立在山脚边，远远看去，如繁星，如瀑布，闪着一片辉煌夺目的亮点儿，一直泻到地上，活像一扇艳丽动人的凤尾，又像一条被舞台灯光照着闪烁发光的长裙。 13、“唧唧啾啾”那是画眉鸟婉转悦耳的叫声，听起来多像歌唱家在唱一首优美动听的歌曲，又像演奏家在吹悠扬悦耳的笛子。 14、层层的叶子中间，零星地点缀着些白花，有袅娜地开着的，有羞涩地打着朵儿的；像一粒粒的明珠，又像碧天里的星星，又像刚出浴的美人。 15、一丝丝红色的花蕊顶着黄色的尖儿探出头来。远看去，一个个小花蕾就像一颗颗珍珠撒在雕刻着花边的红盘中，又像一只只小蜜蜂飞落在花蕊中采蜜。 16、一阵春风拂来，朵朵桃花像一只只花蝴蝶，扇动着美丽地翅膀，翩翩起舞。那些凋谢的花瓣纷纷落下，像仙女散花，又像粉妆玉砌的世界，真叫人赏心悦目，神迷欲醉。 17、抬头望去，深蓝色的天空是那样迷人。空中闪烁着一颗颗明亮的小星星，它们越聚越多，好像在蓝色的地毯上跳舞，又像眨着眼
智能电力仪表的功能及应用
电力SCADA系统的理想选择 智能电力仪表可作为仪表单独使用，取代大量传统的模拟仪表，亦可作为电力监控系统的前端设备，实现远程数据采集与控制。符合工业标准的RS485通讯接口，使得组网轻松便捷，是SCADA系统集成的理想选择。 能量管理系统应用 智能电力仪表可以统计双向四象限的有功电能和无功电能，同时进行最大值/最小值的记录和需量的统计。配合电力监控软件可以协助用户分析各用电设备的电能消耗状况与负荷变化趋势，实现自动抄表并生成各种电量报表。 远程电力控制 SWL300不仅有强大的测量功能，还附带了丰富灵活的I/O功能，这使得它完全可以胜任作为分布式RTU的要求，实现遥信、遥测、遥控、计量于一体。 电能质量监视和分析 SWL300智能电力仪表可以实现在线式的电能质量分析。各相电压、电流的总谐波畸变率（THD），各奇次谐波含有率（3-31次），电压、电流不平衡度以及波峰系数、电话谐波波形因数、K系数均可实时测量。 应用领域 SWL300系列智能电力仪表应用于为电力参数测量、电能质量监视和分析、电气设备控制提供解决方案。 主要应用领域有： ■能源管理系统 ■变电站自动化系统 ■配电自动化系统 ■医院电力监控系统
■工厂自动化系统 ■智能建筑 ■智能型配电盘、开关柜 艾驰商城是国内最专业的MRO工业品网购平台，正品现货、优势价格、迅捷配送，是一站式采购的工业品商城！具有 10年工业用品电子商务领域研究，以强大的信息通道建设的优势，以及依托线下贸易交易市场在工业用品行业上游供应链的整合能力，为广大的用户提供了传感器、图尔克传感器、变频器、断路器、继电器、PLC、工控机、仪器仪表、气缸、五金工具、伺服电机、劳保用品等一系列自动化的工控产品。 如需进一步了解相关仪器仪表产品的选型，报价，采购，参数，图片，批发等信息，请关注艾驰商城https://www.wendangku.net/doc/957662752.html,。
用是什么还是什么造句
用是什么还是什么造句 [标签:栏目] ，用是什么还是什么造句 1、在凄凉秋瑟的细雨中行走，寂寞是指尖的凉和心底的痛。在这冷冷的夜里，寂寞不请自来，穿过肌肤，直抵灵魂深处！抬头望着一望无际的天空，不知是雨水，还是泪水朦胧了我的双眼？ 2、学习不仅是读书、听课、考试那么简单，它还是一种进步的手段，是一种把握人生的能力。 3、鹅毛般的雪花簌簌地不断往下落，织成了天幕雪帘。如同柳絮一般，银一样的白，玉一样的润，一朵朵、一簇簇，纷纷扬扬、冉冉飘落，闪着寒冷的银光。它是天公派来的小天使？还是有人在天上撒下无数透明洁白的梨花瓣。 4、大熊猫是我们的国宝，也是可爱的动物，还是一个活化石。 5、了解一个人，就要看她的全部，不管是晦暗，还是灿烂，都要去周游，都要去看。因为那都是人性的一部分。 6、小草给春天增添了勃勃生机，增添了新的光彩，不管是在贫瘠的土地上，还是在高山上、石缝中，都能见到它翠绿的身影。 7、在我有生之年，只要我眼睛好了，不管她是为人妻还是为人母，她都必是程太太。 8、陈绮贞你在演唱会的时候，为什么要拍手？为什么要站起来跳舞？是因为自己真正受到感动，还是因为跟着别人这样做？回家之后你习惯打开音响，是因为真正喜欢音乐，还是害怕寂寞？ 9、在他的神态和目光里，隐藏着某种混乱、模糊和心不在焉的东西，叫人一看就知道他这个人也许对未来的物质生活，既没有明确的目标，也不怎么关心。可是当他还是一个少年的时候，人们就说过，他是那种想做什么就能把什么做好的人。 10、当你走进餐厅，迎接你的又是一番古色古香的中式格调，即便你知道他家做的是西餐，你还是会感叹于环境的宜人与精致。 11、说，是用嘴来写，无论是一句话，还是一段话，首先要说清楚，想好了再说，把自己要说的话在心里整理一下就能说清楚。
用有时候有时候有时候造句
用有时候有时候有时候造句 导读：1、大海真是变幻莫测，有时候波涛汹涌;有时候波光粼粼;有时候风平浪静。 2、天空中的白云，有时候像飞腾的巨龙，有时候如威武的雄狮，有时候又似奔腾的骏马。 3、有时候月亮像细眉，有时候月亮像小船，有时候月亮像银盘。 4、有时候老师很严厉，有时候老师很和蔼，有时候老师像自己的妈妈。 5、星期天，他有时候打球，有时候上网，有时候逛街，把作业忘到脑后了。 6、小鸟们有时候在天空中展翅高飞，有时候停在树枝上婉转啼叫，有时候在林间欢蹦乱跳。 7、小狗有时候咬骨头;有时候玩玩具;有时候睡大懒觉。 8、天上的白云，有时像温柔的绵羊;有时像奔腾的骏马;有时像调皮的小猫;有时像凶狠的饿狼。 9、月亮有时候象圆盘，有时候像西瓜，有时候象镰刀，有时候又像个光环。 10、我有时候沉默，有时候滔滔不绝，有时候寂寞，有时候调皮，我的生活充满幻想和一大堆愿望…… 11、有时候，生命就像一棵悬崖边的小草，那么孤单;有时候，生命就像一只小虫子，非常脆弱; 12、有时候，生命就像飞机一样，看似慢，其实却很快;有时候，
生命就掌控在自己手中，关键是自己如何去使它绽放光辉! 13、天上的云彩变化多端，有时候像一匹骏马在奔驰，有时候像一条狗在摆头摇尾，有时候像一只鸡在找食，有时候又像一头大象在喝水。 14、人的性情是多变的，有时候高兴的手舞足蹈，有时候伤心的嚎啕大哭，有时候平静的一言不发，有时候又兴奋的喋喋不休。 15、放假时真忙啊，有时候学奥数，有时候练钢琴，有时候学书法，比上学时累多了。 16、人有时像一座永不消融的冰山，有时像一把热情燃烧的火焰，有时像一段丝滑如水的丝绸，有时像闪烁不断的明灯。 17、成长的过程，有时候会有欢乐，有时候会有忧愁，有时候会有迷茫，有时候还会有悲伤：这些其实都是一笔宝贵的财富，因为是我们真实的感受和经历。 18、月亮一个月当中在不停地变化，有时像把镰刀，有时像条小船，有时像圆圆的镜子。 19、有时候天是蓝色的，就像美丽开朗的你如花儿一般绽放;有时候天是墨色的，如喜怒无常的你转眼便是倾盆大雨;有时候天空万里无云，冷冷清清，凄凄惨惨戚戚! 20、有时候，生活就像一杯咖啡，充满了苦涩;有时候，生活就像一杯酒，充满了激情;有时候，生活就像一杯白开水，充满着遐想! 感谢您的阅读，本文如对您有帮助，可下载编辑，谢谢
史上最全的仪表选型 火电厂的你都知道哪些
仪表在工业生产过程中，起着对工艺参数进行检测、显示、记录或控制的重要作用。火电厂需要用到的仪器仪表主要有温度，压力、液位，流量，震动方面的仪器，温度比如热电阻，热电偶、普通压力表，压力变送器，差压变送器、翻板液位计，电极式液位计、孔板，涡街，电磁，巴式等流量计汽机的震动位移探头、DCS系统等。本文为大家介绍自动化仪表、温度仪表、压力仪表、流量仪表、物位仪表的选型，内容非常全面！ 一、自动化仪表选型的一般原则 检测仪表（元件）及控制阀选型的一般原则如下： 1.工艺过程的条件 工艺过程的温度、压力、流量、粘度、腐蚀性、毒性、脉动等因素是决定仪表选型的主要条件，它关系到仪表选用的合理性、仪表的使用寿命及车间的防火、防爆、保安等问题。 2.操作上的重要性 各检测点的参数在操作上的重要性是仪表的指示、记录、积算、报警、控制、遥控等功能选定依据。一般来说，对工艺过程影响不大，但需经常监视的变量，可选指示型；对需要经常了解变化趋势的重要变量，应选记录式；而一些对工艺过程影响较大的，又需随时监控的变量，应设控制；对关系到物料衡算和动力消耗而要求计量或经济核算的变量，宜设积算；一些可能影响生产或安全的变量，宜设报警。 3.经济性和统一性 仪表的选型也决定于投资的规模，应在满足工艺和自控的要求前提下，进行必要的经济核算，取得适宜的性能／价格比。 为便于仪表的维修和管理，在选型时也要注意到仪表的统一性。尽量选用同一系列、同一规格型号及同一生产厂家的产品。 4.仪表的使用和供应情况 选用的仪表应是较为成熟的产品，经现场使用证明性能可靠的；同时要注意到选用的仪表应当是货源供应充沛，不会影响工程的施工进度。 二、温度仪表的选型 <一>一般原则 1单位及标度（刻度） 温度仪表的标度（刻度）单位，统一采用摄氏温度（℃）。 2检出（测）元件插入长度 插入长度的选择应以检出（测）元件插至被测介质温度变化灵敏具有代表性的位置为原则。但在一般情况下，为了便于互换，往往整个装置统一选择一至二挡长度。 在烟道、炉膛及带绝热材料砌体设备上安装时，应按实际需要选用。 检出（测）元件保护套材质不应低于设备或管道材质。如定型产品保护套太薄或不耐腐蚀（如铠装热电偶），应另加保护套管。 安装在易燃易爆场所的就地带电接点的温度仪表、温度开关、温度检出（测）元件和变送器等，应选用防爆型。 <二>就地温度仪表的选型 1精确度等级 一般工业用温度计：选用1.5级或1级。 精密测量和实验室用温度计：应选用0.5级或0.25级。 2测量范围 最高测量值不大于仪表测量范围上限值的90％，正常测量值在仪表测量范围上限值的1／2左右。
用一和就造句
用一和就造句 本文是关于好词好句的，仅供参考，如果觉得很不错，欢迎点评和分享。 用一和就造句 1、他一听到这个消息就逃跑了。 2、我一回家就写作业。 3、从一开始他就坚决反对这项计划。 4、小明一遇到难题，就想问妈妈。 5、雨一停我们就出发。 6、每次他一发出尖细的叫声，她就跳起来看是怎么一回事。 7、她一听到这坏消息就悲伤起来。 8、他一听到喧闹的音乐就兴奋，而我则不然。 9、一想到下雪，我就会想起我的青年时代。 10、你一旦开始一件任务，就必须把它完成。 11、如果你能把这面旗子挂到旗杆顶部，你就可以拿这些奖品中的任何一件。 12、他们就像被塞进罐头一样，挤在一间小房间内。 13、他在第一轮比赛中猛击一拳就把对手打倒在地。 14、我一坐下这件紧身衬衫就从裤子里跑到上面来了。 15、她一听到这消息就立即离开了。 16、用这种快速胶卷，照片仅用一分钟就会显影。
17、你怎么竟敢问也不问我一声就把我的那些旧信扔掉了？ 18、铃声一响，同学们就马上进了教室。 19、她一看到这情况就昏过去了。 20、我一看到上海美丽的夜景，就不想再离开了。 21、那姑娘一看见她母亲来就急忙走开。 22、小明是个贪玩的孩子，他一玩起来就忘记了一切。 23、她一看到油腻的食物就恶心。 24、他一来我们就走。 25、我一不让他喝酒，他就发出“嗬—哼呣”不耐烦的声音。 26、太阳一出来，地上就像下了火。 27、他从一开始就同情那个残疾人。 28、他们一听到这个好消息就非常高兴。 29、车床仅在一分钟内就可车出一根轴。 30、我们一吃完饭，她总是很快就收拾好餐桌。 31、我一听这首伤感的歌曲，就想起那件难过的事情。 32、我一回家就先洗手。 33、敌人一出现，我们就发起了攻击。 34、我一回家就开始写作业。 35、我一上完体育课，就感到口渴。 36、我们一看见这所房子，就把它选作我们的家了。 37、一提到那只狗，她就吓得畏畏缩缩。 38、将这台新电视机与旧电视机一比，你就会看出哪一台更好
一年级用就是造句大全
一年级用就是造句大全 导读：本文是关于一年级用就是造句大全，如果觉得很不错，欢迎点评和分享！ 1、我真喜欢我的学校，我的学校就像花园，我们就像花园里的一朵花，一株草，我们被“园丁”爱护着，那园丁就是我们的老师。 2、如果每一个人都缺乏恒心，没有面对挑战时的坚持，那岂不少了很多发明与创造？人类的进步砌不要慢了很多？坚持就是胜利，这是永远不变的真理！ 3、人们都说日出是最美丽的，日出就是日落，当这里夕阳西下后，地球的另一面就是美丽的黎明。所以我认为日落是最美丽的。因为每临近黄昏时太阳顺着山的那头小心翼翼的走下去，便在人们眼前展现出一幅神秘，美丽而迷人的画。 4、不管多久没见面，我们彼此都还是老样子，脾气差，说话大声，不注意仪表，可是永远笑得那么开心。朋友，就是无论分开生活多久，再相聚也不会尴尬，会为你的伤心而心疼。 5、一株茉莉也许没有沁人心脾的芳香，但它永远会让你感到清新，感到幽雅，父爱就是这样，犹如茉莉一样静静地开放。无论你在何方，父亲那慈爱的眼睛定会伴随你一生。 6、啊，那就是樱花！我一步并两步跑到樱花树前，抬头仔细观赏着。有的繁花满枝，像粉红色的云霞；有的花谢了，树衩上又抽出片片鲜嫩欲滴的树叶；有的羞羞涩涩，把头埋在花苞里说悄悄话呢！
7、江南的天气就是这样，虽热但也凉爽，让人感觉不到那灼灼烈日的炙烤，也无乌云压顶般的闷燥。柳枝摇摆，微风习习，漫步在那河岸边，踩着脚下青石阔道，心灵不觉寂静了下来。摆脱了勾心斗角，远离了喧哗吵闹。 8、如果明天是一幢摩天大厦，今天就是决定那大厦寿命的基石。同学，让我们珍惜今天这一分一秒，把这大厦的基石打得无比坚实。 9、花是我们生活中常见的植物，这个世界因为花而变得多姿多彩。花就是生活，懂得欣赏花的人，也就懂得受享花的鲜艳美丽。 10、月光如水，静静地洒在地上，给大地披上银灰色的纱裙。她穿着白色的衬衫，娴静而安详，温柔而大方，她那玉盘似的脸，透过柳梢，留下温和的笑容，啊，这就是那梦幻般的美吗？ 11、站在这里一看，真怪，山简直变了样，它们的形状与在平原或半山望上来大不相同，它们变得十分层叠、杂乱，雄伟而奇特。往上仰望，山就是天，天也是山，前后左右尽是山，好像你的鼻子都可随时触到山。 12、我理想的生活方式就是花朵的生活方式：自然开放，他人可以观赏和喜欢你，你却不必应酬任何人。我觉得自己从来都是江湖之外的江湖人，因为最初我就是独往独来，现在也还是独往独来。 13、每天不浪费或不虚度或不空抛的那一点点时间，即使只有 五、六分钟，如得正用，也一样可以有很大的成就。游手好闲惯了，就是有聪明才智，也不会有所作为。 14、世界只有一个，就是此刻压迫着你的这个，你也只在这一
用只要就造句优秀句子
用只要就造句优秀句子 只要就是一组关联词，以下是用只要就造句优秀句子，欢迎阅读学习！ 用只要就造句优秀句子【1】⑴只要我们善于把精读和泛读结合起来，就能取访最佳的读书效果。 ⑵只要人人献出份爱心世界会更美妙。 ⑶只要掌握正确的学习方法，学习就会变成得很轻松。 ⑷只要努力学习定会取访优异成绩。 ⑸我们想要的东西，只要是安康有益的，父母就一定会答应我们的。 ⑹只要付出努力，就会有收获。 ⑺只要努力学习，认真听课，就能取访好成绩。 ⑻无论你是学习，还是工作，只要努力，就一定会成功。 ⑼无论什么行业，只要认真努力的做，就可以做好。 ⑽只要你肯付出，就一定会收获回报。 ⑾只要持之以恒，就一定会成功。 ⑿只要我下次努力，就会考好。 1⑶只要贪官污吏把贪的钱拿出来，就能解决中国普通公民普遍供不起孩子上大学问习题。 1⑷只要我们肯付出，就一定会得到回报。 1⑸只要努力学习，就一定会取访好成绩。
1⑹只要人人都献出一点爱，世界就会变得更美妙。 1⑺只要贪图小利踏踏实实工作定会过上幸福生活。 1⑻只要我真诚地对待别人，别人就会真诚地对待我。 1⑼只要自己辛勤努力学习，就一定会取访好成绩。 20、只要认真听课，你就完成了学习的一小半，只要认真完成作业，就离成功不远了，只要把笔记做好，即可以使自己的知识更丰富。 2⑴她的手真巧，只要轻轻一拉，美好的曲子就从她的手指间流淌出来了。 2⑵只要我们能够天天练习，就一定能取访成功。 2⑶只要承诺了，就应该去兑现。 2⑷只要多观察，就会有发现。 2⑸只要有问习题，就一定要解决。 2⑹只要我多做些力所能及家务事妈妈会少我操心点。 2⑺不管是刮风还是下雨，只要我们齐心协力，就一定会到达目的！ 2⑻只要有希望，就应该去努力。 2⑼只要我们努力了，就不会有遗憾。 30、一个人只要活的诚实，就等于有了一大笔财富。 用只要就造句优秀句子【2】⑴只要认真听课，你就完成了学习的一小半，只要认真完成作业，就离成功不远了，只要把笔记做好，便能够使自我的知识更丰富。 ⑵只要努力学习定会取访优异成绩。
仪器仪表
一、基本概念 1、根据曲线进行凑试的控制器参数整定方法称为_________ 。 2、简单均匀控制与简单控制相比，__________。 3、PI控制器积分时间是指在输入作阶跃变化时输出达到纯比例输出的_____ 倍所对应的时间。 4、广义对象包括通常将过程、执行器、检测元件与变送器，不包括控制装置。 5、前馈控制要求主要扰动可测不控可控（开环）。 6、在阶跃扰动作用下过渡过程有衰减振荡等哪些形式。 发散振荡、单调发散、等幅振荡、衰减振荡、单调衰减； 7、设定值保持不变的反馈控制系统称为定值控制系统，其主要任务是：抗拒外界干扰、使被控变量恢复到预期的常量。 8、设定值变化的控制系统称为什么？随动控制系统其主要任务是什么？输出（指示值）应迅速和正确的随着输入值而变化（测量值）。 9、差压式流量计利用节流件前后的__________ 和__________的关系测得流量。 10、被控变量是指什么，操纵变量是指什么，偏差是指什么，被控对象是指什么？ 被控变量：被控过程内要求保持设定值的工艺参数。操作变量：受控制器操纵的用以克服干扰的影响，使被控变量保持设定值的物料量或能量。偏差：被控变量实际值与设定值之差。被控对象：自动控制系统中，工艺参数要满足要求需要控制的生产过程、设备或机器。 11、希望控制通道的放大系数怎么样，时间常数怎么样，时滞怎么样？k0稍大、T0适中、希望τo小 12、希望扰动通道的放大系数怎么样，时间常数怎么样，时滞怎么样？Kf小、希望Tf大、希望τc大 13、热电偶参比端温度将低，测得热电势数值将怎样变化？使用补偿导线的目的是什么？。 4、孔板流量计压差增加一倍，则体积流量为原来的多少倍？密度增加一倍，则体积流量为原来的多少倍？15．下列有关控制阀理想流量特性的表述，不正确的是_________ 。 （1）阀门前后压差固定（2）阀门前后压差为0 （3）特性取决于结构参数（4）线性阀放大系数是常数 16．一般马路上交通信号红绿灯的周期变化控制属于哪种控制？程序控制系统设定值也是变化的，但它是一个已知的时间函数，即根据需要按一定时间程序变化。 17、涡轮流量计、涡街流量计、椭圆齿轮流量计、电磁流量计，哪种属于容积式流量计？ 8、双闭环控制系统可以用两个简单控制系统取代。 19、前馈控制属于开环控制还是闭环控制？
用就是和也造句大全
用就是和也造句大全 导读：本文是关于用就是和也造句大全，如果觉得很不错，欢迎点评和分享！ 1、我喜欢宁静的生活。严格地说，我觉得自己从来都是江湖之外的江湖人。因为最初我就是独往独来，现在也还是独往独来。 2、一个人有了很大的理想，就是在最艰苦困难的时候，也会感到幸福。 3、东北风呜呜地叫着。枯草落叶满天飞扬，黄尘蒙蒙，混沌一片，简直分辨不出何处是天，何处是地了。就是骄傲的大鹰，也不敢在这样的天气里，试试它的翅膀。 4、这么多的油菜花，一朵有一朵的姿势，看看这一朵，很美；看看那一朵也很美。如果把眼前的一片油菜花看做一大幅美丽的图画，我就是这片油菜花中的一朵，我也要把自己的芬芳散发出来。 5、这个世界本身就是不公平的，地球仪也永远是倾斜的转着。 6、我家饭桌上没有哪一样是父亲的私菜，小孩更不许吃独食，不许霸占好菜，不许独占座位。大起来有些才能的人最大的毛病就是霸气，世界不太平也是因为霸气，实在可思省。 7、我突然明白，就是以一个傻子的眼光来看，这个世界也不是完美无缺的。这个世界上任何东西都是这样，你不要它，它就好好地在那里，保持着它的完整，它的纯粹；一旦到了手中，你就会发现，自己没有全部得到。
8、成功是白天的太阳，那么失败就是黑夜中的星辰，没有星辰的降落也就不会有太阳的升起，耀眼的太阳也会有被乌云遮掉的时候。 9、作为父母，让孩子在十五六岁就远离家乡，远离亲人，只身到外面去求学深造，当然是有些于心不忍，但是为了他们的将来，就是再不忍心也要忍心。 10、到下蛋的时候，它差不多是发了狂，恨不能让全世界都知道它这点成绩，就是聋子也会被吵得受不了。 11、我们现在最重要的任务就是学习，也是我们也是我们一生都得追求的目标。 12、这只狗还非常负责任，经常把眼睛挣得大大的，探测着有没有小偷，就是有外人给他扔食物他也不会吃，照样昂首挺胸的探测。天天守着家门口，只有主人让他去玩，他才出去，从来不乱跑。 13、真理就是火炬，同时也是一支巨大的火把。因此我们都想眯着眼睛走过去，甚至担心会被烫伤。 14、节约时间，也就是使一个人的生命更加有效，也就等于延长了人的生命。 15、做成功一个店之后离你大的成功就不远了，所以你首先就是脚踏实地、集中精力地先做出一家，也是要放弃掉连锁的这种，不要在将来如何做连锁方面做太多的梦，先脚踏实地做出第一家。 16、职场的人际关系就是简单的交易关系，不管你是老板也好、打工仔也好，你在职场中所遇到的困难和烦恼都是因为你和别人的期望不一致、有冲突造成的。
一年级用又造句大全
一年级用又造句大全 导读：本文是关于一年级用又造句大全，如果觉得很不错，欢迎点评和分享！ 1、那手也不是我所记得的红活圆实的手，却又粗又笨而且开裂，像是松树皮了。 2、太阳公公露出了久违而又可爱的笑容，将温暖的阳光洒向大地，花儿抬头张开花瓣，开出了美丽的花朵。 3、春天像个爱帮助别人的小天使，它让树换上新衣，它又让小树长的又高又壮。 4、法赫米这个人既开朗又含蓄，既善于交际而又喜欢磨人。 5、春天，山楂树上开满了美丽、洁白而又明媚的小花，在绿色叶子的衬托下，显得格外美丽、娇艳。 6、她的小胳膊又黄又瘦，一点血色也没有，那上面还有大小不一的褐色斑块。 7、沙滩长长的，像一条玉带，踩在上面又松又软，又像是银缎地毯。 8、孩子是一张白纸，纯洁而又干净，随着时间的描绘，最终变成一幅色彩斑斓的图画。 9、啊，春！她迈着轻盈的步伐而又含着少女的羞涩不敢大步走来，她悄悄撕裂了灰蒙蒙的冬天的外衣，离我们越来越近。 10、下午我回家一看，一只又大又黄中带绿的柚子安安稳稳的
摆放在桌子上。一看它那"胖胖"的样子，就让我口水直流。 11、既能美化环境，又能提醒大家要爱护环境。 12、墙壁上映出一个高高的身影，两条腿又细又长，支撑还没有充分发育的上身，像一只重心不稳的高脚杯。 13、漫天的雨纷然而又漠然，广不可及的灰色中竟有这样一株红莲！像一堆即将燃起的火，像一罐立刻要倾泼的颜色！ 14、抬头一看，一只又大又肥的知了正趴在树干上拼命地叫着。 15、她浅笑盈盈，眸子像浸在水里的墨晶石，又黑又亮。 16、我长着一双又黑又大的眼睛，大家都叫我大眼睛姑娘。我是一个活泼阳光的女孩。 17、世界上最快而又最慢，最长而又最短，最平凡而又最珍贵，最容易被忽视而又最令人后悔的，就是时间。 18、我看了蜘蛛侠的电影，真是又惊心动魄又精彩啊！ 19、小于连既聪明又勇敢，用智慧和勇气救了全城的老百姓。 20、妈妈又要工作，又忙家务，任劳任怨，从不叫苦。 21、她既学习好又守纪律，既美丽又善良。 22、正如一粒粒的明珠，又如碧天里的星星，又如刚出浴的美人。 23、看到多年未见的好友，我感到又惊又喜。 24、小白兔吃起东西来，三瓣嘴一动一动的，很有趣。它的前肢短，后肢长，走起路来总是一蹦一跳，跑得又快又远。 25、她又瘦又长，发青的脸，发青的脖子，站在竖着旗杆的高
智能电力仪表市场分析报告
智能电力仪表市场分析报告 综述: 智能网络多功能电力仪表是针对电力系统、工矿企业、公共设施、智能大厦的电力监控需求而设计的。它能测量所有的常用电力参数，如三相电流、电压，有功、无功功率，电度、谐波等。由于该电力仪表还具备完善的通信联网功能，所以我们称之为网络多功能电力仪表。智能网络多功能电力仪表具有很高的性能价格比，可以直接取代常规电力变送器及测量仪表。作为一种先进的智能化、数字化的前端采集元件，该系列网络仪表已广泛应用于各种控制系统（如：SCADA数据采集与监视控制系统、IPDS智能配电系统和EMS能源管理系统）中。 智能电网与物联网的提出，给智能电力仪表带来了新的发展与机遇。用户端的智能化建设从通常意义上讲主要分为智能配电系统、能源管理系统、智能楼宇系统，这三大系统的建设都离不开智能电力仪表。在不同的应用场合，功能可以有不同的组合，大致分三类仪表，即监控与保护仪表、电能分析与管理仪表、电气安全仪表。 1 国内市场发展状况 智能电力仪表在2000年左右进入中国市场，初期主要在电力、石化等高端用户使用，替代传统指针表和电量变送器，产品主要由国外著名电气公司提供。随着用户对用电可靠、安全、节约的要求提高，配用电系统智能化也越来越普及，从而推动了智能电力仪表的应用，市场从原有的国外品牌一统天下，到目前国内外多品牌竞争的格局。我国市场上国外著名品牌主要以施耐德、西门子、溯高美为主，自主品牌主要为斯菲尔、安科瑞、珠海派诺等。其创立时间、主营产品以及市场定位见表1。 表1 电力仪表著名公司创立日期、主营产品及市场定位
器为66万台，塑料外壳式断路器为2900万台，由于整个行业企业近千家，规模较大的已统计在内，整个行业的实际数据应在上述数据的3倍左右，因此，与开关配套的电工仪表也应在6000万台左右。
用是还是造句
用是还是造句 他不仅是一个音乐家，还是一个画家。 对于书本要求我们是一味执行，还是该适当改革。 老师是我们的知识源泉，还是我们的心灵启迪站和灵魂的工程师。 1、我相信，真正在乎我的人是不会被别人抢走的，无论是友情，还是爱情。 2、书是最好的朋友。唯一的缺点是使我近视加深但还是值得的。 3、人只有一双眼，只能望着前面。所以，不管做什么事情，不管是得意还是失意，都要记得看看身后。百密仍有一疏，而这一输往往决定胜败。 4、在他的神态和目光里，隐藏着某种混乱、模糊和心不在焉的东西，叫人一看就知道他这个人也许对未来的物质生活，既没有明确的目标，也不怎么关心。可是当他还是一个少年的时候，人们就说过，他是那种想做什么就能把什么做好的人。 5、小丑是那种无论心里是开心还是难过别人都看不出来的人，因为他给自己画上了笑脸。 6、见过很多类型的男人后，最终觉得男人最大的特点就是单纯，即便年纪大了也还是像小孩子的感觉。如果一个男的总是让女友感到他的成熟，那么，我想，这个女人可能没有能走进他的内心。
7、该支持的还是会支持，该厌恶的还是会厌恶，两类人我都感谢，因为无论是怎样都是一种坚持的纯粹。 8、我不在乎是悲伤的离别还是不痛快的离别，只要是离开一个地方，我总希望离开的时候自己心中有数。 9、时间到底是善良的，还是邪恶的魔术师?都不是。时间只是一种简单的乘法，使原来的数值倍增而已。 10、人的一生总有注定错失的因缘，和你携手相伴的人或许不是你要的那杯茶，但你还是要强忍着苦涩饮下。所以彼此厌倦并不是谁的过错，只怪造化弄人，无端生出这么多的痴男怨女，不得尽如人愿。 11、我记住的蘸着此生所有失望写下的温柔。你一定会幸福的，可这幸福不是我给的想起来还是会很难过。 12、我唯有继续前行，去寻找那极乐的净土。然而那净土不在人间，我不知这是勇气还是逃避。 13、我不知道他喜欢怎样的姑娘，我一直只想给他看最好的摸样，却时时不能如愿，让他觉得任性，觉得我只是个小孩子。明明是个没有心的死人，还是会觉得悲伤，我不知道该怎么办。 14、你爱过的那些人，在起初貌似完美无缺。当他们逐渐四分五裂变成一堆碎片，你是否仍能用掌心托起和保存。你爱的是他人的属性，还是他们的面具和形式。 15、像和是之间，还是距离得太遥远了。 16、在我有生之年，只要我眼睛好了，不管她是为人妻还是为
只要就造句
只要就造句 本文是关于好词好句的，仅供参考，如果觉得很不错，欢迎点评和分享。 只要就造句 1、只要人人都献出一点爱，世界就会变得更美好。 2、只要努力学习就能获得知识同时也能考取好的大学。 3、只要努力学习定会取得优异成绩。 4、只要我多做些力所能及家务事妈妈会少我操心点。 5、只要我下次努力，就会考好。 6、只要努力学习，认真听课，就能取得好成绩。 7、只要有一点爱，一点情，一点关怀，便是友谊的花盛开的地方，就能播友谊之种，开友谊之花，繁殖友谊之花园。 8、只要努力学习，就一定会取得好成绩。 9、一个人只要活的诚实，就等于有了一大笔财富。 10、只要掌握正确的学习方法，学习就会变成得很轻松。 11、只要我们肯付出，就一定会得到回报。 12、只要开始了，就不轻言放弃。 13、只要自己辛勤努力学习，就一定会取得好成绩。 14、只要付出努力，就会有收获。 15、只要你肯付出，就一定会收获回报。 16、只要有希望，就不能放弃。
17、只要我们善于把精读和泛读结合起来，就能取得最佳的读书效果。 18、她的手真巧，只要轻轻一拉，美妙的曲子就从她的手指间流淌出来了。 19、只要我真诚地对待别人，别人就会真诚地对待我。 20、只要勇敢面对，就会战胜一切困难。 21、只要人人献出份爱心世界会更美好。 22、不管是刮风还是下雨，只要我们齐心协力，就一定会达到目标！ 23、无论你是学习，还是工作，只要努力，就一定会成功。 24、只要承诺了，就应该去兑现。 25、只要贪图小利踏踏实实工作定会过上幸福生活。 26、只要有希望，就应该去努力。 27、只要多观察，就会有发现。 28、只要我们能够天天练习，就一定能取得成功。 29、只要明天不下雨，我们就去春游。 30、只要我们努力了，就不会有遗憾。 31、无论什么行业，只要认真努力的做，就可以做好。 32、只要坚持不懈，就一定会成功。 33、只要有问题，就一定要解决。 感谢阅读，希望能帮助您！
用是什么还是什么造句
用是什么还是什么造句 导读：本文是关于用是什么还是什么造句，如果觉得很不错，欢迎点评和分享！ 1、小丑是那种无论心里是开心还是难过别人都看不出来的人，因为他给自己画上了笑脸。 2、我唯有继续前行，去寻找那极乐的净土。然而那净土不在人间，我不知这是勇气还是逃避。 3、家，是外出游子不变的港湾，无需任何证件我们就可以在那里停驻，无论是和平年代还是**年代。落叶归根，也是我们不变的宿命。 4、番茄，他的外表是红彤彤的，像一个个小灯笼，可爱极了。它含有大量的维生素和蛋白质。它的果肉不管是生着吃、还是煮着吃，味道都一样香甜可口。 5、窗台摆放着十多盆兰花，那兰花浓郁的香气，黄绿的花瓣，那深绿色细长简洁的叶片，不知觉中给我很深的印象，这是花还是草呢！ 6、不论你是一个男子还是一个女子，待人温和宽大才配得上人的名称。一个人的真正的英勇果敢，决不等于用拳头制止别人发言。 7、不管是风吹，还是雨打；也不管是骄阳的暴晒，小草都不畏艰难，毅然决然的挺了下来，小草那种坚强的品质深深的打动了我的心。
8、没错可是无论你做什么，这个种子最终也还是长成棵桃树，你或许想要苹果或是橘子，但到头来还是桃子。 9、陈绮贞你在演唱会的时候，为什么要拍手？为什么要站起来跳舞？是因为自己真正受到感动，还是因为跟着别人这样做？回家之后你习惯打开音响，是因为真正喜欢音乐，还是害怕寂寞？ 10、我的父亲很高，但很瘦。他的背已有些伛偻，我不知道是他更瘦了还是老了。这是一个让人思考的问题，想得我揪心地疼。 11、秋天，是丰收的，还是忧伤的？一切都要看你自己的体会了。 12、不断学习，不管你长得如何，不管你是不是有钱，不管是这样还是那样，你都是有机会的。 13、如果你想要自己幸福，那么请用心去爱别人。我们生存的世界，不仅是用五彩缤纷的事物构成的，还是用爱编织的。当你全心全意的爱一个人时，你会感到无比幸福。 14、一面面的雨帘不断交替，雨点瞬间从天空落下时，速度出奇得快，不假思索的就落下来，或许人有时就得像雨点那样想做就做，不需要磨磨蹭蹭，不需要想自己是落在地上还是被鲜花树叶接住。 15、遇到这样的事，真不知道是该羞愧还是愤怒，但是仔细想来，还是努力提升自己最重要。 16、我好喜欢这个文具盒哦！因为它不仅是我学习上的好帮手，还是生活中的好伙伴。有了它，我的学习成绩一定会更加优秀！ 17、微风拂面，那像手一般的抚摸，带给我们无限的遐想，是