TRUETIME 1.5：自动控制领域参考手册——功能与技术详解

3星 · 超过75%的资源需积分: 5 134 浏览量更新于2024-08-02 收藏 1.6MB PDF 举报

TRUETIME 1.5 Reference Manual 是一份由 Martin Ohlin、Dan Henriksson 和 Anton Cervin 所编写的指南，出自 Lund University 自动控制部门，发布于2007年1月。该手册详细介绍了 TRUETIME 1.5 软件系统，这是一款专为实时仿真和嵌入式控制设计的工具包，适用于Matlab和C++编程环境。该手册分为多个部分： 1. **介绍**：阐述了TRUETIME的背景和目的，强调其在实时模拟中的应用，特别是对于硬件和软件同步的精确性需求。 2. **入门指南**： - **软件要求**：指出了使用TRUETIME 1.5所需的最低软件配置，包括对Matlab和C++的支持。 - **安装**：提供了安装步骤，包括系统配置和依赖项安装。 - **编译**：讲解如何编译TRUETIME以适应特定平台。 3. **使用模拟器**：详细说明如何在TRUETIME环境中创建和操作模型，可能涉及模拟硬件网络行为如CSMA/CD（以太网）和CAN（控制器局域网）。 4. **编写代码函数**： - **Matlab代码函数**：指导用户如何编写与TRUETIME集成的Matlab函数，用于处理算法或数据处理。 - **C++代码函数**：同样介绍了在C++中实现代码功能的方法，强调跨平台的兼容性和性能优化。 - **调用Simulink块图**：解释如何将Simulink模型与TRUETIME交互，进行实时仿真。 5. **初始化**：这部分关注如何设置系统的初始状态，包括编写Matlab和C++脚本以初始化模型参数。 6. **编译**：深入讨论编译过程中的注意事项，可能涉及针对不同硬件架构的优化策略。 7. **TrueTime内核**：重点介绍TrueTime kernel，它负责实现精确的时间同步和调度，部分章节讲解了动态电压调整技术。 8. **TrueTime网络**：涵盖多种网络通信模型，如轮询令牌总线（TokenBus）、频分多址（FDMA）、时分多址（TDMA，如TTP）以及无线网络（如802.11b/g Wi-Fi和802.15.4 ZigBee）的实现。 9. **TrueTime无线网络**：详述了无线通信在TRUETIME中的应用场景，包括错误概率计算和自定义路径损耗函数。 10. **TrueTime电池管理**：探讨电池供电的实时应用，可能涉及能量效率和电源管理算法。 11. **独立网络模块**：介绍了TRUETIME中预置的独立网络模块，这些模块可以作为用户项目的补充。 12. **示例**：手册中包含实际案例和代码示例，以便读者学习和理解如何在TRUETIME环境中实施各种控制和通信策略，如PID控制。 TRUETIME 1.5 Reference Manual 是一个全面的文档，涵盖了从系统安装、编程接口到高级特性的详细介绍，对于从事实时系统开发和嵌入式控制的工程师来说，是不可或缺的参考资源。

Figure 6 The dialog of the TRUETIME network block.

The network block is configured through the block mask dialog, see Figure 6.

Using the command ttSetNetworkParameter, see Section 14, it is also possible to

change some parameters on a per node basis. The following network parameters

are common to all models:

Network number The number of the network block. The networks must be

numbered from 1 and upwards. Wired and wireless networks are not al-

lowed to use the same number.

Number of nodes The number of nodes that are connected to the network.

This number will determine the size of the Snd, Rcv and Schedule input

and outputs of the block.

Data rate (bits/s) The speed of the network.

Minimum frame size (bits) A message or frame shorter than this will be

padded to give the minimum length. Denotes the minimum frame siz e,

including any overhead introduced by the protocol. E.g., the minimum Eth-

ernet frame size, including a 14-byte header and a 4-byte CRC, is 512 bits.

Pre-processing delay (s) The time a message is delayed by the network in-

terface on the sending end. This can be used to model, e.g ., a slow serial

connection between the computer and the network interface.

Post-processing delay (s) The time a message is delayed by the network in-

terface on the receiving end.

Loss probability (0 –1) The probability that a network message is lost during

transmission. Lost messages will consume network bandwidth, but will

never arrive at the destination.

8.1 CSMA/CD (Ethernet)

CSMA/CD stands for Carrier Sense Multiple Access with Collision Detection. If

the network is busy, the sender will wait until it occurs to be free. A collision

will occur if a m essage is transmitted within 1 microsecond of another (this

corresponds to the propagation delay in a 200 m cable; the actual number is not

very important since collisions are only likely to occur when two or more nodes

are waiting for the cable to be idle). W hen a collision occurs, the sender will back

of f for a time defined by

backoff

= minimum frame size / data rate $ R

where R = rand(0, 2

−1) (discrete uniform distribution) and K is the number

of collisions in a row (but maximum 10—there is no upper limit on the number of

retransmissions, however). Note that for CSMA/CD, minimum frame size cannot

be 0.

After waiting, the node will attempt to retransmit. In an example where two

nodes are waiting for a third node to finish its transmission, they will first

collide with p robability 1, then with probability 1/2 (K = 1), then 1/4 (K = 2),

and so on.

8.2 CSMA/AMP (CAN)

CSMA/AMP stands for Carrier Sense Multiple Access with Arbitration on Mes-

sage Priority. If the network is busy, the sender will wait until it occurs to be

free. If a collision occurs (again, if two transmissions are being started within

1 microsecond), the message with the highest pri ority (the lowest priority num-

ber) will continue to be transmitted. If two m essages w i th the same priority

seek transmission simultaneously, an arbitrary choice is made as to which is

transmitted first. (In real CAN applications, all sending nodes have a unique

identifier, which serves as the message priority.)

8.3 Round Robin (Token Bus)

The nodes in the network take turns (from lowest to highest node number) to

transmit one frame each. Between turns, the network is idle for a time

idle

= minimum frame size / date rate,

representing the time to pass a token to the next node.

8.4 FDMA

FDMA stands for Frequency Division Multiple Access. The transmissions of the

different nodes are completely independent and no collisions can occur. In this

mode, there is an extra attribute

Bandwidth allocations A vector of shares for the sender nodes which must

sum to at most one.

The actual bit rate of a sender is computed as (allocated bandwidth $ data rate).

8.5 TDMA (TTP)

TDMA stands for Time Division Multiple Access. Works similar to FDMA, except

that each node has 100 % of the bandwidth but only in its scheduled slots. If a

full frame cannot be transmitt ed in a slot, the transmission will continue in the

next scheduled slot, without any extra penalty. Note that overhead i s added to

each frame just as in the other protocols. The extra attributes are

Slot size (bits) The size of a sending slot. The slot time is hence given by

slot

= slot size / data rate.

Schedule A vector of sender node ID’s (1 . . . nbrOfNodes) specifyi ng a cyclic

send schedule. A zero is also an allowed node ID, meaning that no-one is

allowed to transmit in that time slot.

8.6 Switched Ethernet

In Switched Ethernet, each node in the network has its own, full-duplex con-

nection to a central switch. Compared to an ordinary Ethernet, there will never

be any collisions on the network segments in a Switched Ethernet. The switch

stores the received messages in a buffer and then forwards them to the correct

destination nodes. This common scheme is known as store and forward.

If many messages in the switch are destined for the same node, they are trans-

mitted in FIFO order. There can be either one queue that holds all the messages

in the switch, or one queue for each output segment. In case of heavy traffic and

long message queues, the switch may run out of memory. The following options

are associated with the Switched Ethernet:

Total switch memor y (bits) This is the total amount of memory available for

storing messages in the switch. An amount of memory equal to the length

of the message is allocated when the message has been fully received in

the switch. The same memory is deallocated when the complete message

has reached its final destination node.

Switch buffer type This setting describes how the memory is allocated in the

switch. Comm on buffer means that all messages are stored in a single FIFO

queue and share the same memory area. Symmetric output buffers means

that the memory is divided into n equal parts, one for each output segment

connected t o the switch. When one output queue runs out of memory, no

more messages can be stored in that particular queue.

Switch overflow behavior This option describes what happens when the

switch has run out of memory. When the complete message has been re-

ceived in the switch, it is deleted. Retransmit means that the switch then

informs the sending node that it should try to retransmit the message.

Drop means that no notification is given—the message is simply deleted.

Network number The number of the network block. The networks must be

numbered from 1 and upwards. Wired and wireless networks are not al-

lowed to use the same number.

Number of nodes The number of nodes that are connected to the network.

This number will determine the size of the Snd, Rcv and Schedule input

and outputs of the block.

Data rate (bits/s) The speed of the network.

Minimum frame size (bits) A message or frame shorter than this will be

padded to give the minimum length. Denotes the minimum frame siz e,

including any overhead introduced by the protocol. E.g., most network pro-

tocols have a fixed number of header and tail bits, so the f rame must be at

least s izeo f (header)+ sizeo f (tail) long.

Tran smit power Determines how strong the radio signal will be, and thereby

how long it will reach.

Receiver signal threshold If the received energy is above this threshold, then

the medium is accounted as busy.

Path-loss exponent The path loss of the radio signal is modeled as

where

d is the distance in meters and a is a suitably chosen parameter to model

the environment. Typically chosen in the interval 2-4.

ACK timeout The time a sending node will wait for an ACK message before

concluding that the message was lost and retransmit it.

Retr y limit The maximum number of times a node will try to retransmit a

message before giving up.

Error coding threshold A number in the interval [0, 1] which defines the per-

centage of block errors in a message that the coding can handle. For exam-

ple, cer tain coding schemes can fully reconstruct a message if it has less

than 3% block errors. The number of block errors are calculated using the

signal-to-noise ratio, where the noise is all other ongoing transmissions.

9.1 80 2 .11b/g (W LAN)

IEEE 802.11b/g is used in many laptops and mobile devices of today. The protocol

is based on CSMA/CA with some modifications.

In the simulation, a package transmission is modeled li ke this: The node that

wants to transmit a packet checks to see i f the medium i s idle. The transmission

may proceed, if the medium is found to be idle, and has stayed so for 50

s. If,

on the other hand, the medium is found to be busy, a random back-off time is

chosen and decrement ed in the same way as when colliding (described later in

this section). When a node starts to transmit, its relative position to all other

nodes in the same network is calculated, and the signal level in all those nodes

are calculated according to the path-loss formula

The signal is assumed to be possible to detect if the signal level in the receiving

node is larger than the receiver signal threshold. If this is the case, then the

signal-to-noise ratio (SNR) is calculated and used to fi nd the block error rate

(BLER). Note that all other transmissions add to the background noise when

calculating the SNR. The BLER, together with the size of the message, is used

to calculate the number of bit errors in the message and if the percentage of

剩余106页未读，继续阅读

shanpohe

粉丝: 11
资源: 23

TRUETIME 1.5：自动控制领域参考手册——功能与技术详解

最新资源