Title: ISHM spring 2005 Communication Protocols
1ISHM spring 2005Communication Protocols
2Contents
- What are the needs of protocols to support ISHM
- Protocols that apply to ISHM
- Protocols that are best suitable for ISHM
3What needs must be addressed by new protocols
to be suitable to support ISHMfunctionality?
Reliable Fast Ease of integration between
systems with different communication
requirements (legacy systems) Fault
tolerant Support large of nodes Time
synchronization support Standardization of
health information communication
4- Low error rate
- Variable packet size
- Reconfigurable
- Speed/bandwidth
- Standard changes with system level
5standards exist that apply to ISHM system?
- Synchronous
- 1553
- 1773
- RS485
- RS232
- NTP (Network time protocol)
- TTP (Time triggered protocol)
- PTP (Point to point protocol)
6Asynchronous
- 1. Ethernet (Data link protocols)
- 2. 1451
-
- 3. CAN (Controller area network)
-
- TCP/IP (Communication Protocols)
- UDP (User datagram protocol)
-
7Three Standard Protocols that apply to ISHM
- TTP Time Triggered Protocol
- CAN controller area network
- TCP/IP Transmission Control Protocol/Internet
- protocol
8Time Triggered Protocol
- The time-triggered protocol TTP/C is based on
time as its underlying driving force, i.e., all
activities of a system are carried out in
response to the passage of certain points in time - It guaranteed that no single failure of any part
of the communication system could lead to a
disrupture of the communication
9Features of the TTP
- Supports fault-tolerance
- Autonomy and Composability
- Error detection
- Integrates numerous services
- Predictable message transmission
- Message acknowledgement in group communication
- Clock synchronization
- Redundancy management
- Temporary blackout handling
10Structure of the TTP/C communication controller.
11System Overview
- FTU- single or replicated nodes
- Replicated communication channels
- The channel is a broadcast bus
- Local nodes time synchronized by TTP
12System Configurations
- Class I
- One node per FTU
- 2 frames per FTU
- Used when cable failures are dominant
- Frame
- A bit packet that is transported on the physical
level - May contain one or more messages
13System Configurations
- Class II
- Two active nodes per FTU
- 2 frames per FTU
14System Configurations
- Class III
- Two node per FTU
- 4 frames per FTU
15Comparison of Configurations
16Protocol Description
- Frame format
- Initialization frame- used during start-up.
C-state sent in data field - Normal frame
17Protocol Description
- CRC
- Every node stores a membership vector and the
global time -- C-state. - membership vector has a bit position for each
node-24 nodes/cluster - Each frame Xmits a CRC.
- CRC is calc. from user data
- If 2 different nodes have C-state then CRC is not
correct.. Reset C state. - Fault-tolerant average algorithm applied to
maintain continuous clock synchronization
18Protocol Description
- Redundancy management and initialization
- Some nodes send I-frame periodically, so new
nodes may reintegrate - A new node sends out its C-state in I-frame until
it recognizes a frame from another node - If all sending slots are occupied, it is a shadow
node - Temporary blackout handling
- Correlated failure of a number of nodes
- Identified by sudden drop in membership
- Nodes send I-messages and perform local emergency
control - After membership has stabilized, mode changed to
global emergency service
19In Conclusion- Strengths
- Only time-triggered architectures can provide
fault-tolerant real-time performance - The TTP is efficient and scalable
- Can be implemented using available hardware,
signaling mechanisms - Low overhead
- High data rates, used in both twisted fiber and
optical channels - Reusability, composability, and testability
20In Conclusion- Weaknesses
- The schedule is fixed so there is no bandwidth
allocated for alarms and other spontaneous
messages - All fault-tolerance mechanism is implemented at
system level, this means that very little
freedom is left for application specific
implementations - Expensive
- Addition of nodes affects the existing system
(although not the application)
21Controller Area Network Protocol
- Broadcast comm. Mechanism i.e. Message oriented
transmission - Every message has message identifiers
- They are unique in the whole network.
- They define the content and priority of the
message. - This feature makes the system configuration more
flexible, easy to add changes to the network,
allows multiple data processing.
22CAN-Protocol
Frame
Filter
Filter
23Real-time data transmission
Listen only
Listen only
24Message format
Start of frame Identifier Remote transmission
request Identifier extension Data length
code Data field Cyclic redundancy
check Acknowledge End of frame Intermission
frame space
25Detecting and signalling errors
- Cyclic Redundancy Check (CRC)
- FCS (frame check sequence) is added at
the transmission ends and recomputed - Frame check
- compares with the original format.
- ACK errors
- receivers acknowledge ACK frame
26The CAN protocol also implements two mechanisms
for error detection at the bit level
- Monitoring
- Each station that transmits also observes
the bus level - Bit stuffing
- five consecutive equal bits the transmitter
inserts a stuff bit into the bit stream
27Trade-offs of CAN
- The bus latency time is longer for extended frame
format. - require more bandwidth
28UDP
- Connection less protocol
- It is just IP with a short header added
- UDP provides a way for application to send
encapsulated IP datagrams - Send them without having to establish a
connection.
29Applications of UDP
- Video conferencing
- Internet radio
- Other real-time multimedia applications
30UDP Header
32 bits
Source port
Destination port
UDP checksum
UDP length
31Real-time Transport Protocol
- The position of RTP in the protocol stack.
Socket interface
32RTP Payload
UDP payload
IP payload
Ethernet payload
- The basic function of RTP is to multiplex several
real time data streams into single stream of UDP
packets
33- Set Baud Rate menu
- The Baud rate is 230,400
- Bit-error rate
- bit-error rate of at most 1e-5
34Advantages
- it doesn't restrict you to a connection based
communication model, so startup latency in
distributed applications is much lower - all flow control, tracking, transaction logging,
etc is up to user programs. - broadcast and multicast transmission are
available with udp.
35Disadvantages
- there are no guarantees with udp. a packet may
not be delivered, or delivered twice, or
delivered out of order. - udp has no flow control. implementation is the
duty of user programs. - routers are quite careless with udp. they never
retransmit it if it collides, udp suffers from
worse packet loss than tcp.
36TCP
- It is designed to provide a reliable end-to-end
byte stream over an unreliable internetwork. - It provides reliable, sequential delivery of
data.
37TCP header
38TCP connection management
39Performance
- TCP causes the entire stream from any lost
portion to be retransmitted in its entirety - In high bit-error-rate (BER) scenarios this will
cause large amounts of bandwidth to be wasted in
resending data that has already been successfully
received
40Advantages
- TCP guarantees three things
- that your data gets there,
- that it gets there in order,
- and that it gets there without duplication.
- routers may notice tcp packets and treat them
specially. they can buffer and retransmit them, - TCP has good relative throughput on a modem or a
LAN.
41Disadvantages
- tcp may have lots of features we don't need. it
may waste bandwidth, time, or effort on ensuring
things that are irrelevant to the task at hand. - tcp has no block boundaries we must create our
own. - routers on the internet today are out of memory.
design assumptions of tcp break down in this
environment. - tcp has relatively poor throughput on a lossy,
high bandwidth, high latency link, such as a
satellite connection or an overfull t1.
42Questions